ADSP-21267SKBCZ-X [ADI]

Preliminary Technical Data; 初步的技术数据
ADSP-21267SKBCZ-X
型号: ADSP-21267SKBCZ-X
厂家: ADI    ADI
描述:

Preliminary Technical Data
初步的技术数据

微控制器和处理器 外围集成电路 数字信号处理器 时钟
文件: 总44页 (文件大小:486K)
中文:  中文翻译
下载:  下载PDF数据表文档文件
SHARC® Processor  
ADSP-21267  
Preliminary Technical Data  
DAI incorporates two precision clock generators (PCG), and  
an input data port (IDP) that includes a parallel data acqui-  
sition port (PDAP), and three programmable timers, all  
under software control by the signal routing unit (SRU)  
On-chip memory—1M Bit of on-chip SRAM and a dedicated  
3M Bits of on-chip mask-programmable ROM  
The ADSP-21267 is available with a 150 MHz core instruction  
rate. For complete ordering information, see Ordering  
Guide on page 43  
SUMMARY  
High performance 32-bit/40-bit floating point processor  
optimized for high performance audio processing  
Code compatible with all other SHARC DSPs  
The ADSP-21267 processes high performance audio while  
enabling low system costs  
Audio decoders and post processor-algorithms support.  
Non-volatile memory can be configured to contain a com-  
bination of PCM 96 kHz, Dolby Digital, Dolby Digital EX2,  
Dolby Pro Logic IIx, DTS 5.1, DTS ES Discrete 6.1, DTS-ES  
Matrix 6.1, DTS Neo:6, MPEG2x BC (2 channel) and others.  
See www.analog.com/SHARC for a complete list  
Single-Instruction Multiple-Data (SIMD) computational archi-  
tecture—two 32-bit IEEE floating-point/32-bit fixed-point/  
40-bit extended precision floating point computational  
units, each with a multiplier, ALU, shifter, and register file  
High bandwidth I/O —a parallel port, an SPI port, four serial  
ports, a digital audio interface (DAI) and JTAG test port  
Figure 1. FUNCTIONAL BLOCK DIAGRAM  
CORE PROCESSOR  
DUAL PORTED MEMORY  
BLOCK 0  
DUAL P ORT ED ME MORY  
BLOCK 1  
INSTRUCTION  
CACHE  
32 X 48-BIT  
SRAM  
0.5 MBIT  
SRAM  
0.5 MBIT  
TIMER  
ROM  
ROM  
1.5 MBIT  
1.5 MBIT  
DAG1  
8X4X32  
DAG2  
8X4X32  
PROGRAM  
SEQUENCER  
ADDR  
DATA  
ADDR  
DATA  
32  
32  
PM ADDRESS BUS  
DM ADDRESS BUS  
64  
64  
PM DATA BUS  
DM DATA BUS  
IOD  
(32)  
IOA  
(1 8)  
DMA CONTROLLER  
22 CHANNELS  
PX REGISTER  
4
GPIO FLAGS/  
IRQ/TIMEXP  
PROCESSING  
ELEMENT  
(PEX)  
PROCESSING  
ELEMENT  
(PEY)  
4
SPI PORT (1)  
ADDRE SS/  
D ATA BUS/ GP IO  
16  
3
6
CON TR OL/G PIO  
PARALLEL  
PORT  
SERIAL PORTS (6)  
JTAG TEST & EMULATION  
IOP  
REGISTERS  
(MEMORY MAPPED)  
SIGNAL  
ROUTING  
UNIT  
INPUT  
20  
DATA PORTS (8)  
PARALLEL DATA  
ACQUISITION PORT  
CONTROL,  
STATUS, &  
DAT A BUFFERS  
PRECISION CLOCK  
GENERATORS (2)  
S
3
TI ME RS (3)  
DIGITAL AUDIO INTERFACE  
I/O PROCESSOR  
SHARC and the SHARC logo are registered trademarks of Analog Devices, Inc.  
Rev. PrA  
Information furnished by Analog Devices is believed to be accurate and reliable.  
However, no responsibility is assumed by Analog Devices for its use, nor for any  
infringements of patents or other rights of third parties that may result from its use.  
Specifications subject to change without notice. No license is granted by implication  
or otherwise under any patent or patent rights of Analog Devices. Trademarks and  
registered trademarks are the property of their respective owners.  
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106 U.S.A.  
Tel:781/329-4700 www.analog.com  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Serial Ports provide:  
KEY FEATURES  
Four dual data line serial ports that operate at 37.5M Bits/s  
on each data line —each has a clock, frame sync and two  
data lines that can be configured as either a receiver or  
transmitter pair  
Left-justified Sample Pair and I2S Support, programmable  
direction for up to 16 simultaneous receive or transmit  
channels using two I2S compatible stereo devices per  
serial port  
TDM support for telecommunications interfaces including  
128 TDM channel support for newer telephony inter-  
faces such as H.100/H.110  
Up to 4 full-duplex TDM streams, each with 128 channels  
per frame  
Companding selection on a per channel basis in TDM mode  
Input Data Port provides an additional input path to the DSP  
core configurable as either eight channels of I2S or serial  
data or as seven channels plus a single 20-bit wide syn-  
chronous parallel data acquisition port  
Supports receive audio channel data in I2S, Left-justified  
sample pair, or right-justified mode  
Signal Routing Unit (SRU) provides configurable and flexible  
connections between all DAI components, four serial  
ports, three timers, 10 interrupts, six flag inputs, six flag  
outputs, two precision clock generators, an input data  
port/parallel data acquisition port, and 20 SRU I/O pins  
(DAI_Px)  
At 150 MHz (6.65 ns) core instruction rate, the ADSP-21267  
operates at 900 MFLOPS performance whether operating  
on fixed or floating point data  
300 MMACS sustained performance at 150 MHz  
Code compatibility—At assembly level, uses the same  
instruction set as other SHARC DSPs  
Super Harvard Architecture—three independent buses for  
dual data fetch, instruction fetch, and nonintrusive, zero-  
overhead I/O  
1M Bit on-chip dual-ported SRAM (0.5M Bit in block 0 and  
0.5M Bit in block 1) for simultaneous access by core proces-  
sor and DMA  
3M Bits on-chip dual-ported mask-programmable ROM (1.5M  
Bits in block 0 and 1.5M Bits in block 1)  
Dual Data Address Generators (DAGs) with modulo and bit-  
reverse addressing  
Zero-overhead looping with single-cycle loop setup, provid-  
ing efficient program sequencing  
Single Instruction Multiple Data (SIMD) architecture  
provides:  
Two computational processing elements  
Concurrent execution— Each processing element executes  
the same instruction, but operates on different data  
DMA Controller supports:  
18 zero-overhead DMA channels for transfers between  
ADSP-21267 internal memory and the four serial ports,  
the input data port (IDP) , SPI-compatible port, and the  
parallel port  
Serial Peripheral Interface (SPI)  
Master or slave serial boot through SPI  
Full-duplex operation  
32-bit background DMA transfers at core clock speed, in  
parallel with full-speed processor execution  
Asynchronous parallel/external port provides:  
Access to asynchronous external memory  
16 multiplexed address/data lines that can support 24-bit  
address external address range with 8-bit data or 16-bit  
address external address range with 16-bit data  
50 Mbyte per sec transfer rate  
Master-Slave mode multi-master support  
Open drain outputs  
Programmable baud rates, clock polarities and phases  
3 Muxed Flag/IRQ lines  
1 Muxed Flag/Timer expired line  
ROM Based Security features:  
JTAG access to memory permitted with a 64-bit key  
Protected memory regions that can be assigned to limit  
access under program control to sensitive code  
PLL has a wide variety of software and hardware multi-  
plier/divider ratios  
JTAG background telemetry for enhanced emulation  
features  
IEEE 1149.1 JTAG standard test access port and on-chip  
emulation  
256 word page boundaries  
External memory access in a dedicated DMA channel  
8- to 32- bit and 16- to 32-bit word packing options  
Programmable wait state options: 2 to 31 CCLK  
Digital Audio Interface (DAI) includes four serial ports, two  
precision clock generators, an input data port/parallel data  
acquisition port, three timers and a signal routing unit  
Dual voltage: 3.3 V I/O, 1.2 V core  
Available in 136-ball BGA and 144-lead LQFP packages  
Also available in lead-free packages  
Rev. PrA  
|
Page 2 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
GENERAL DESCRIPTION  
The ADSP-21267 SHARC DSP is a member of the SIMD  
SHARC family of DSPs featuring Analog Devices' Super Har-  
vard Architecture. The ADSP-21267 is source code compatible  
with the ADSP-2136x, and ADSP-2116x DSPs as well as with  
first generation ADSP-2106x SHARC processors in SISD (Sin-  
gle-Instruction, Single-Data) mode. Like other SHARC DSPs,  
the ADSP-21267 is a 32-bit/40-bit floating-point processor opti-  
mized for high performance audio applications with its dual-  
ported on-chip SRAM, mask-programmable ROM, multiple  
internal buses to eliminate I/O bottlenecks, and an innovative  
Digital Audio Interface (DAI).  
• Three Programmable Interval Timers with PWM Genera-  
tion, PWM Capture/Pulse width Measurement, and  
External Event Counter Capabilities  
• On-Chip dual-ported SRAM (1 Mbit)  
• On-Chip dual-ported, mask-programmable ROM  
(3 Mbits)  
• JTAG test access port  
• 8- or 16-bit Parallel port that supports interfaces to off-chip  
memory peripherals  
• DMA controller  
As shown in the Functional Block Diagram on page 1, the  
ADSP-21267 uses two computational units to deliver a signifi-  
cant performance increase over previous SHARC processors on  
a range of DSP algorithms. Fabricated in a state-of-the-art, high  
speed, CMOS process, the ADSP-21267 DSP achieves an  
instruction cycle time of 6.6 ns at 150 MHz. With its SIMD  
computational hardware, the ADSP-21267 can perform 900  
MFLOPS running at 150 MHz.  
• Four full-duplex serial ports  
• SPI-compatible interface  
• Digital Audio Interface that includes two precision clock  
generators (PCG), an input data port (IDP), four serial  
ports, eight serial interfaces, a 20-bit synchronous parallel  
input port, 10 interrupts, six flag outputs, six flag inputs,  
three timers, and a flexible signal routing unit (SRU)  
Table 1 shows performance benchmarks for the ADSP-21267.  
Figure 2 on page 4 shows one sample configuration of a SPORT  
using the precision clock generator to interface with an I2S ADC  
and an I2S DAC with a much lower jitter clock than the serial  
port would generate itself. Many other SRU configurations are  
possible.  
Table 1. ADSP-21267 Benchmarks (at 150 MHz)  
Benchmark Algorithm  
Speed  
(at 150 MHz)  
1024 Point Complex FFT (Radix 4, with  
reversal)  
FIR Filter (per tap)1  
61.3 µs  
ADSP-21267 FAMILY CORE ARCHITECTURE  
The ADSP-21267 is code compatible at the assembly level with  
the ADSP-2136x, ADSP-2116x, and with the first generation  
ADSP-2106x SHARC DSPs. The ADSP-21267 shares architec-  
tural features with the ADSP-2136x and ADSP-2116x SIMD  
SHARC family of DSPs, as detailed in the following sections.  
3.3 ns  
IIR Filter (per biquad)1  
13.3 ns  
Matrix Multiply (pipelined)  
[3x3] x [3x1]  
[4x4] x [4x1]  
30 ns  
53.3 ns  
SIMD Computational Engine  
Divide (y/x)  
20 ns  
30 ns  
The ADSP-21267 contains two computational processing ele-  
ments that operate as a Single-Instruction Multiple-Data  
(SIMD) engine. The processing elements are referred to as PEX  
and PEY and each contains an ALU, multiplier, shifter and reg-  
ister file. PEX is always active, and PEY may be enabled by  
setting the PEYEN mode bit in the MODE1 register. When this  
mode is enabled, the same instruction is executed in both pro-  
cessing elements, but each processing element operates on  
different data. This architecture is efficient at executing math  
intensive audio algorithms.  
Inverse Square Root  
1 Assumes two files in multichannel SIMD mode.  
The ADSP-21267 continues SHARC’s industry leading stan-  
dards of integration for DSPs, combining a high performance  
32-bit DSP core with integrated, on-chip system features. These  
features include 1M bit dual-ported SRAM memory, 3M bits  
dual-ported ROM, an I/O processor that supports 18 DMA  
channels, four serial ports, an SPI interface, an external parallel  
bus, and Digital Audio Interface (DAI).  
Entering SIMD mode also has an effect on the way data is trans-  
ferred between memory and the processing elements. When in  
SIMD mode, twice the data bandwidth is required to sustain  
computational operation in the processing elements. Because of  
this requirement, entering SIMD mode also doubles the band-  
width between memory and the processing elements. When  
using the DAGs to transfer data in SIMD mode, two data values  
are transferred with each access of memory or the register file.  
The block diagram of the ADSP-21267 on page 1, illustrates the  
following architectural features:  
• Two processing elements, each containing an ALU, Multi-  
plier, Shifter and Data Register File  
• Data Address Generators (DAG1, DAG2)  
• Program sequencer with instruction cache  
• PM and DM buses capable of supporting four 32-bit data  
transfers between memory and the core at every core pro-  
cessor cycle  
Rev. PrA  
|
Page 3 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
ADSP-21267  
CLKOUT  
CLKIN  
XTAL  
CLOCK  
ALE  
2
CLK_CFG1-0  
LATCH  
AD15-0  
ADDR  
PARALLEL  
PORT  
RAM, ROM  
BOO T ROM  
I/O DEVICE  
2
3
BOOTCFG1-0  
FLG3-1  
DATA  
OE  
RD  
WR  
WE  
FLG0  
CS  
ADC  
(OPTIONAL)  
CLK  
FS  
DAI_P1  
DAI_P2  
DAI_P3  
SDAT  
SCLK0  
SFS0  
SRU  
SD0A  
SD0B  
DAC  
(OPTIONAL)  
CLK  
DAI_P18  
DAI_P19  
DAI_P20  
SPORT0  
SPORT1  
SPORT2  
SPORT 3  
FS  
SDAT  
CLK  
F S  
PCGA  
PCGB  
DAI  
RESET  
JTAG  
6
Figure 2. ADSP-21267 System Sample Configuration  
Independent, Parallel Computation Units  
Single-Cycle Fetch of Instruction and Four Operands  
Within each processing element is a set of computational units.  
The computational units consist of an arithmetic/logic unit  
(ALU), multiplier and shifter. These units perform all opera-  
tions in a single cycle. The three units within each processing  
element are arranged in parallel, maximizing computational  
throughput. Single multi-function instructions execute parallel  
ALU and multiplier operations. In SIMD mode, the parallel  
ALU and multiplier operations occur in both processing ele-  
ments. These computation units support IEEE 32-bit single-  
precision floating-point, 40-bit extended precision floating-  
point, and 32-bit fixed-point data formats.  
The ADSP-21267 features an enhanced Harvard architecture in  
which the data memory (DM) bus transfers data and the pro-  
gram memory (PM) bus transfers both instructions and data  
(see the Figure 1 on page 1). With the ADSP-21267’s separate  
program and data memory buses and on-chip instruction cache,  
the processor can simultaneously fetch four operands (two over  
each data bus) and one instruction (from the cache), all in a sin-  
gle cycle.  
Instruction Cache  
TheADSP-21267 includes an on-chip instruction cache that  
enables three-bus operation for fetching an instruction and four  
data values. The cache is selective—only the instructions whose  
fetches conflict with PM bus data accesses are cached. This  
cache allows full-speed execution of core, looped operations  
such as digital filter multiply-accumulates and FFT butterfly  
processing.  
Data Register File  
A general purpose data register file is contained in each process-  
ing element. The register files transfer data between the  
computation units and the data buses, and store intermediate  
results. These 10-port, 32-register (16 primary, 16 secondary)  
register files, combined with the ADSP-2126x enhanced Har-  
vard architecture, allow unconstrained data flow between  
computation units and internal memory. The registers in PEX  
are referred to as R0-R15 and in PEY as S0-S15.  
Data Address Generators With Zero-Overhead Hardware  
Circular Buffer Support  
The ADSP-21267’s two data address generators (DAGs) are  
used for indirect addressing and implementing circular data  
buffers in hardware. Circular buffers allow efficient program-  
ming of delay lines and other data structures required in digital  
Rev. PrA  
|
Page 4 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
signal processing, and are commonly used in digital filters and  
Fourier transforms. The two DAGs of the ADSP-21267 contain  
sufficient registers to allow the creation of up to 32 circular buff-  
ers (16 primary register sets, 16 secondary). The DAGs  
automatically handle address pointer wrap-around, reduce  
overhead, increase performance, and simplify implementation.  
Circular buffers can start and end at any memory location.  
the parallel port. Eighteen channels of DMA are available on the  
ADSP-21267 — one for the SPI interface, eight via the serial  
ports, eight via the Input Data Port and one via the processor’s  
parallel port. Programs can be downloaded to the ADSP-21267  
using DMA transfers. Other DMA features include interrupt  
generation upon completion of DMA transfers, and DMA  
chaining for automatic linked DMA transfers.  
Flexible Instruction Set  
Digital Audio Interface (DAI)  
The 48-bit instruction word accommodates a variety of parallel  
operations, for concise programming. For example, the ADSP-  
21267 can conditionally execute a multiply, an add, and a sub-  
tract in both processing elements while branching and fetching  
up to four 32-bit values from memory; all in a single instruction.  
The Digital Audio Interface (DAI) provides the ability to con-  
nect various peripherals to any of the DSPs DAI pins  
(DAI_P[20:1]).  
Programs make these connections using the Signal Routing  
Unit (SRU, shown in the block diagram on page 1).  
ADSP-21267 MEMORY AND I/O INTERFACE  
FEATURES  
The SRU is a matrix routing unit (or group of multiplexers) that  
enables the peripherals provided by the DAI to be intercon-  
nected under software control. This allows easy use of the DAI  
associated peripherals for a much wider variety of applications  
by using a larger set of algorithms than is possible with non-  
configurable signal paths.  
The ADSP-21267 adds the following architectural features to  
the SIMD SHARC family core:  
Dual-Ported On-Chip Memory  
The DAI also includes 4 serial ports, 2 precision clock genera-  
tors (PCG), an input data port (IDP), 6 flag outputs and 6 flag  
inputs, and 3 timers. The IDP provides an additional input path  
to the ADSP-21267 core, configurable as either eight channels  
of I2S or serial data or as seven channels plus a single 20-bit wide  
synchronous parallel data acquisition port Each data channel  
has its own DMA channel that is independent from the ADSP-  
21267's serial ports.  
The ADSP-21267 contains one megabit of internal SRAM and  
three megabits of internal mask-programmable ROM. Each  
block can be configured for different combinations of code and  
data storage (see ADSP-21267 Memory Map on page 6). Each  
memory block is dual-ported for single-cycle, independent  
accesses by the core processor and I/O processor. The dual-  
ported memory, in combination with three separate on-chip  
buses, allow two data transfers from the core and one from the  
I/O processor, in a single cycle.  
For complete information on using the DAI, see the ADSP-  
2126x SHARC DSP Peripherals Manual.  
On the ADSP-21267, the SRAM can be configured as a maxi-  
mum of 32K words of 32-bit data, 64K words of 16-bit data, 21K  
words of 48-bit instructions (or 40-bit data), or combinations of  
different word sizes up to one megabit. All of the memory can  
be accessed as 16-bit, 32-bit, 48-bit, or 64-bit words. A 16-bit  
floating-point storage format is supported that effectively dou-  
bles the amount of data that may be stored on-chip. Conversion  
between the 32-bit floating-point and 16-bit floating-point for-  
mats is performed in a single instruction. While each memory  
block can store combinations of code and data, accesses are  
most efficient when one block stores data using the DM bus for  
transfers, and the other block stores instructions and data using  
the PM bus for transfers.  
Serial Ports  
The ADSP-21267 features four full duplex synchronous serial  
ports that provide an inexpensive interface to a wide variety of  
digital and mixed-signal peripheral devices such as the AD183x  
family of audio codecs, ADCs, and DACs. The serial ports are  
made up of two data lines, a clock and frame sync. The data  
lines can be programmed to either transmit or receive and each  
data line has its own dedicated DMA channel.  
Serial ports are enabled via 8 programmable and simultaneous  
receive or transmit pins that support up to 16 transmit or 16  
receive channels of audio data when all four SPORTS are  
enabled, or four full duplex TDM streams of 128 channels per  
frame.  
Using the DM bus and PM buses, with one dedicated to each  
memory block assures single-cycle execution with two data  
transfers. In this case, the instruction must be available in the  
cache.  
The serial ports operate at up to one-quarter of the DSP core  
clock rate, providing each with a maximum data rate of 37.5  
Mbits/s for a 150 MHz core. Serial port data can be automati-  
cally transferred to and from on-chip memory via a dedicated  
DMA. Each of the serial ports can work in conjunction with  
another serial port to provide TDM support. One SPORT pro-  
vides two transmit signals while the other SPORT provides the  
two receive signals. The frame sync and clock are shared.  
DMA Controller  
The ADSP-21267’s on-chip DMA controller allows zero-over-  
head data transfers without processor intervention. The DMA  
controller operates independently and invisibly to the processor  
core, allowing DMA operations to occur while the core is simul-  
taneously executing its program instructions. DMA transfers  
can occur between the ADSP-21267’s internal memory and its  
serial ports, the SPI-compatible (serial peripheral interface)  
port, the IDP (input data port/parallel data acquisition port) or  
Serial ports operate in four modes:  
• Standard DSP serial mode  
• Multichannel (TDM) mode  
Rev. PrA  
|
Page 5 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
INTERNAL MEMORY  
SPACE  
LONG WORD  
ADDRESSING  
NORMAL WORD  
ADDRESSING  
SHORT WORD  
ADDRESSING  
IOP REGISTERS  
IOP REGISTERS  
IOP REGISTERS  
0x0000 0000 - 0x0003 FFFF  
0x0000 0000 - 0x0003 FFFF  
0x0000 0000 - 0x0003 FFFF  
BLOCK 0 SRAM (0.5 Mbit)  
0x0004 0000 - 0x0004 1FFF  
BLOCK 0 SRAM (0.5 Mbit)  
0x0008 0000 - 0x0008 3FFF  
BLOCK 0 SRAM (0.5 Mbit)  
0x0010 0000 - 0x0010 7FFF  
RESERVED  
RESERVED  
RESERVED  
0x0004 2000 - 0x0005 7FFF  
0x0008 4000 - 0x000A FFFF  
0x0010 8000 - 0x0015 FFFF  
BLOCK 0 ROM (1.5 mbit)  
0x0005 8000 - 0x0002 FFFF  
BLOCK 0 ROM (1.5 mbit)  
0x000B 0000 - 0x000B BFFF  
BLOCK 0 ROM (1.5 mbit)  
0x0016 0000 - 0x0017 7FFF  
RESERVED  
RESERVED  
RESERVED  
0x0005 3000 - 0x0005 FFFF  
0x000B C000 - 0x000B FFFF  
0x0017 8FFF - 0x0017 FFFF  
BLOCK 1 SRAM (0.5 Mbit)  
0x0006 0000 - 0x0006 1FFF  
BLOCK 1 SRAM (0.5 Mbit)  
0x000C 0000 - 0x000C 3FFF  
BLOCK 1 SRAM (0.5 Mbit)  
0x0018 0000 - 0x0018 7FFF  
RESERVED  
RESERVED  
RESERVED  
0x0006 2000 - 0x0007 7FFF  
0x000C 4000 - 0x000E FFFF  
0x0018 8000 - 0x001D FFFF  
BLOCK 1 ROM (1.5 mbit)  
0x0007 8000 - 0x0007 DFFF  
BLOCK 1 ROM (1.5 mbit)  
0x000F 0000 - 0x000F BFFF  
BLOCK 1 ROM (1.5 mbit)  
0x001E 0000 - 0x001F 7FFF  
RESERVED  
0x0007 E000 - 0x0007 FFFF  
RESERVED  
0x000F C000 - 0x000F FFFF  
RESERVED  
0x000  
EXTERNAL MEMORY  
SPACE  
1
EXTERNAL MEMORY IS NOT DIRECTLY ACCESSIBLE  
BY THE CORE. DMA MUST BE USED TO READ OR WRITE  
TO THIS MEMORY USING THE SPI OR PARALLEL PORT.  
RESERVED  
0x0020 0000 - 0x00FF FFFF  
2
BLOCK 0 ROM HAS A 48-BIT ADDRESS RANGE  
(0x000A 0000 - 0x000A AAAA).  
EXTERNAL DMA  
3
1
BLOCK 1 ROM HAS A 48-BIT ADDRESS RANGE  
ADDRESS SPACE  
0x0100 0000 - 0x02FF FFFF  
(0x000E 0000 - 0x000E AAA).  
RESERVED  
0x0300 0000 - 0x3FFF FFFF  
Figure 3. ADSP-21267 Memory Map  
• I2S mode  
32 bits. For the Left-justified Sample Pair and I2S modes, data-  
word lengths are selectable between 8 bits and 32 bits. Serial  
ports offer selectable synchronization and transmit modes as  
well as optional µ-law or A-law companding selection on a per  
channel basis. Serial port clocks and frame syncs can be inter-  
nally or externally generated.  
• Left-justified sample pair mode  
Left-justified Sample Pair Mode is a mode where in each Frame  
Sync cycle two samples of data are transmitted/received — one  
sample on the high segment of the frame sync, the other on the  
low segment of the frame sync. Programs have control over var-  
ious attributes of this mode.  
Serial Peripheral (Compatible) Interface  
Each of the serial ports supports the Left-justified Sample Pair  
and I2S protocols (I2S is an industry standard interface com-  
monly used by audio codecs, ADCs and DACs such as the  
Analog Devices AD183x family), with two data pins, allowing  
four Left-justified Sample Pair or I2S channels (using two stereo  
devices) per serial port, with a maximum of up to 16 audio  
channels. The serial ports permit little-endian or big-endian  
transmission formats and word lengths selectable from 3 bits to  
Serial Peripheral Interface (SPI) is an industry standard syn-  
chronous serial link, enabling the ADSP-21267 SPI-compatible  
port to communicate with other SPI-compatible devices. SPI is  
an interface consisting of two data pins, one device select pin,  
and one clock pin. It is a full-duplex synchronous serial inter-  
face, supporting both master and slave modes. The SPI port can  
operate in a multi-master environment by interfacing with up to  
four other SPI-compatible devices, either acting as a master or  
Rev. PrA  
|
Page 6 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
slave device. The ADSP-21267 SPI-compatible peripheral  
Phased Locked Loop  
implementation also features programmable baud rates up to  
37.5 MHz, clock phases, and polarities. The ADSP-21267 SPI-  
compatible port uses open drain drivers to support a multi-mas-  
ter configuration and to avoid data contention.  
The ADSP-21267 uses an on-chip Phase Locked Loop (PLL) to  
generate the internal clock for the core. On power up, the  
CLKCFG1-0 pins are used to select ratios of 16:1, 8:1, and 3:1.  
After booting, numerous other ratios can be selected via soft-  
ware control. The ratios are made up of software configurable  
numerator values from 1 to 32 and software configurable divi-  
sor values of 1, 2, 4, 8, and 16.  
Parallel Port  
The Parallel Port provides interfaces to SRAM and peripheral  
devices. The multiplexed address and data pins (AD15-0) can  
access 8-bit devices with up to 24 bits of address, or 16-bit  
devices with up to 16 bits of address. In either mode, 8- or 16-  
bit, the maximum data transfer rate is one-third the core clock  
speed. As an example, for a clock rate of 150 MHz, this is equiv-  
alent to 50 Mbytes/sec.  
Power Supplies  
The ADSP-21267 has separate power supply connections for the  
internal (VDDINT), external (VDDEXT), and analog (AVDD/AVSS  
)
power supplies. The internal and analog supplies must meet the  
1.2 V requirement. The external supply must meet the 3.3 V  
requirement. All external supply pins must be connected to the  
same power supply.  
DMA transfers are used to move data to and from internal  
memory. Access to the core is also facilitated through the paral-  
lel port register read/write functions. The RD, WR, and ALE  
(Address Latch Enable) pins are the control pins for the parallel  
port.  
Note that the analog supply (AVDD) powers the ADSP-21267’s  
clock generator PLL. To produce a stable clock, you should pro-  
vide an external circuit to filter the power input to the AVDD pin.  
Place the filter as close as possible to the pin. For an example cir-  
cuit, see Figure 4. To prevent noise coupling, use a wide trace  
for the analog ground (AVSS) signal and install a decoupling  
capacitor as close as possible to the pin. Note that the AVSS and  
Timers  
The ADSP-21267 has a total of four timers: a core timer able to  
generate periodic software interrupts and three general purpose  
timers that can that can generate periodic interrupts and be  
independently set to operate in one of three modes:  
A
VDD pins specified in Figure 4 are inputs to the SHARC and not  
the analog ground plane on the board.  
• Pulse Waveform Generation mode  
• Pulse Width Count/Capture mode  
• External Event Watchdog mode  
10  
VDDINT  
AVDD  
0.1F  
0.01F  
The core timer can be configured to use FLAG3 as a Timer  
Expired output signal, and each general purpose timer has one  
bidirectional pin and four registers that implement its mode of  
operation: a 6-bit configuration register, a 32-bit count register,  
a 32-bit period register, and a 32-bit pulse width register. A sin-  
gle control and status register enables or disables all three  
general purpose timers independently.  
AVSS  
Figure 4. Analog Power (AVDD) Filter Circuit  
TARGET BOARD JTAG EMULATOR CONNECTOR  
ROM Based Security  
Analog Devices DSP Tools product line of JTAG emulators uses  
the IEEE 1149.1 JTAG test access port of the ADSP-21267 pro-  
cessor to monitor and control the target board processor during  
emulation. Analog Devices DSP Tools product line of JTAG  
emulators provides emulation at full processor speed, allowing  
inspection and modification of memory, registers, and proces-  
sor stacks. The processor's JTAG interface ensures that the  
emulator will not affect target system loading or timing.  
The ADSP-21267 has a ROM security feature that provides  
hardware support for securing user software code by preventing  
unauthorized reading from the internal code when enabled.  
When using this feature, the DSP does not boot-load any exter-  
nal code, executing exclusively from internal SRAM/ROM.  
Additionally, the DSP is not freely accessible via the JTAG port.  
Instead, a unique 64-bit key, which must be scanned in through  
the JTAG or Test Access Port, will be assigned to each customer.  
The device will ignore a wrong key. Emulation features and  
external boot modes are only available after the correct key is  
scanned.  
For complete information on Analog Devices’ SHARC DSP  
Tools product line of JTAG emulator operation, see the appro-  
priate emulator hardware user's guide.  
DEVELOPMENT TOOLS  
Program Booting  
The ADSP-21267 is supported by a complete automotive refer-  
ence design and development board as well as by a complete  
home audio reference design board available from Analog  
Devices. These boards implement complete audio decoding and  
post processing algorithms that are factory programmed into  
The internal memory of the ADSP-21267 boots at system  
power-up from an 8-bit EPROM via the parallel port, an SPI  
master, an SPI slave or an internal boot. Booting is determined  
by the Boot Configuration (BOOTCFG1-0) pins. Selection of  
the boot source is controlled via the SPI as either a master or  
slave device, or it can immediately begin executing from ROM.  
Rev. PrA  
|
Page 7 of 44  
|
January 2004  
 
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
the ROM of the ADSP-21267. SIMD optimized libraries con-  
sume less processing resources, which results in more available  
processing power for custom proprietary features.  
• Perform source level debugging  
• Create custom debugger windows  
The VisualDSP++ IDDE lets programmers define and manage  
DSP software development. Its dialog boxes and property pages  
let programmers configure and manage all of the SHARC devel-  
opment tools, including the color syntax highlighting in the  
VisualDSP++ editor. This capability permits programmers to:  
The non-volatile memory of the ADSP-21267 can be configured  
to contain a combination of PCM 96 KHz, Dolby Digital, Dolby  
Digital EX2, Dolby Pro Logic IIx, DTS 5.1, DTS Matrix 6.1, DTS  
Discrete 6.1, DTS Neo:6, and MPEG2 2 channel.  
Multiple S/PDIF and analog I/Os are provided to maximize end  
system flexibility.  
• Control how the development tools process inputs and  
generate outputs  
The ADSP-21267 is supported with a complete set of  
CROSSCORE™ software and hardware development tools,  
including Analog Devices emulators and VisualDSP++™ devel-  
opment environment. The same emulator hardware that  
supports other SHARC processors also fully emulates the  
ADSP-21267.  
• Maintain a one-to-one correspondence with the tool’s  
command line switches  
The VisualDSP++ Kernel (VDK) incorporates scheduling and  
resource management tailored specifically to address the mem-  
ory and timing constraints of DSP programming. These  
capabilities enable engineers to develop code more effectively,  
eliminating the need to start from the very beginning, when  
developing new application code. The VDK features include  
Threads, Critical and Unscheduled regions, Semaphores,  
Events, and Device flags. The VDK also supports Priority-based,  
Preemptive, Cooperative, and Time-Sliced scheduling  
The VisualDSP++ project management environment lets pro-  
grammers develop and debug an application. This environment  
includes an easy to use assembler (which is based on an alge-  
braic syntax), an archiver (librarian/library builder), a linker, a  
loader, a cycle-accurate instruction-level simulator, a C/C++  
compiler, and a C/C++ runtime library that includes DSP and  
mathematical functions. A key point for these tools is C/C++  
code efficiency. The compiler has been developed for efficient  
translation of C/C++ code to DSP assembly. The SHARC has  
architectural features that improve the efficiency of compiled  
C/C++ code.  
approaches. In addition, the VDK was designed to be scalable. If  
the application does not use a specific feature, the support code  
for that feature is excluded from the target system.  
Because the VDK is a library, a developer can decide whether to  
use it or not. The VDK is integrated into the VisualDSP++  
development environment, but can also be used via standard  
command line tools. When the VDK is used, the development  
environment assists the developer with many error-prone tasks  
and assists in managing system resources, automating the gen-  
eration of various VDK based objects, and visualizing the  
system state, when debugging an application that uses the VDK.  
The VisualDSP++ debugger has a number of important fea-  
tures. Data visualization is enhanced by a plotting package that  
offers a significant level of flexibility. This graphical representa-  
tion of user data enables the programmer to quickly determine  
the performance of an algorithm. As algorithms grow in com-  
plexity, this capability can have increasing significance on the  
designer’s development schedule, increasing productivity. Sta-  
tistical profiling enables the programmer to non intrusively poll  
the processor as it is running the program. This feature, unique  
to VisualDSP++, enables the software developer to passively  
gather important code execution metrics without interrupting  
the real-time characteristics of the program. Essentially, the  
developer can identify bottlenecks in software quickly and effi-  
ciently. By using the profiler, the programmer can focus on  
those areas in the program that impact performance and take  
corrective action.  
VisualDSP++ Component Software Engineering (VCSE) is  
Analog Devices technology for creating, using, and reusing soft-  
ware components (independent modules of substantial  
functionality) to quickly and reliably assemble software applica-  
tions. Download components from the Web and drop them into  
the application. Publish component archives from within Visu-  
alDSP++. VCSE supports component implementation in  
C/C++ or assembly language.  
Use the Expert Linker to visually manipulate the placement of  
code and data on the embedded system. View memory utiliza-  
tion in a color-coded graphical form, easily move code and data  
to different areas of the DSP or external memory with the drag  
of the mouse, examine run time stack and heap usage. The  
Expert Linker is fully compatible with existing Linker Definition  
File (LDF), allowing the developer to move between the graphi-  
cal and textual environments.  
Debugging both C/C++ and assembly programs with the  
VisualDSP++ debugger, programmers can:  
• View mixed C/C++ and assembly code (interleaved source  
and object information)  
• Insert breakpoints  
• Set conditional breakpoints on registers, memory,  
and stacks  
In addition to the software and hardware development tools  
available from Analog Devices, third parties provide a wide  
range of tools supporting the SHARC processor family. Hard-  
ware tools include SHARC processor PC plug-in cards. Third  
party software tools include DSP libraries, real-time operating  
systems, and block diagram design tools.  
• Trace instruction execution  
• Perform linear or statistical profiling of program execution  
• Fill, dump, and graphically plot the contents of memory  
Rev. PrA  
|
Page 8 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
DESIGNING AN EMULATOR-COMPATIBLE DSP  
BOARD (TARGET)  
The Analog Devices family of emulators are tools that every  
DSP developer needs to test and debug hardware and software  
systems. Analog Devices has supplied an IEEE 1149.1 JTAG  
Test Access Port (TAP) on each JTAG DSP. Nonintrusive in-  
circuit emulation is assured by the use of the processor’s JTAG  
interface—the emulator does not affect target system loading or  
timing. The emulator uses the TAP to access the internal fea-  
tures of the DSP, allowing the developer to load code, set  
breakpoints, observe variables, observe memory, and examine  
registers. The DSP must be halted to send data and commands,  
but once an operation has been completed by the emulator, the  
DSP system is set running at full speed with no impact on sys-  
tem timing.  
To use these emulators, the target board must include a header  
that connects the DSP’s JTAG port to the emulator.  
For details on target board design issues including mechanical  
layout, single processor connections, multiprocessor scan  
chains, signal buffering, signal termination, and emulator pod  
logic, see the EE-68: Analog Devices TAG Emulation Technical  
Reference on the Analog Devices website (www.analog.com).  
Use site search on “EE-68.” This document is updated regularly  
to keep pace with improvements to emulator support.  
ADDITIONAL INFORMATION  
This data sheet provides a general overview of the ADSP-21267  
architecture and functionality. For detailed information on the  
ADSP-2126x Family core architecture and instruction set, refer  
to the ADSP-2126x DSP Core Manual and the ADSP-21160  
SHARC DSP Instruction Set Reference.  
Rev. PrA  
|
Page 9 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
PIN FUNCTION DESCRIPTIONS  
ADSP-21267 pin definitions are listed below. Inputs identified  
as synchronous (S) must meet timing requirements with respect  
to CLKIN (or with respect to TCK for TMS, TDI). Inputs iden-  
tified as asynchronous (A) can be asserted asynchronously to  
CLKIN (or to TCK for TRST). Tie or pull unused inputs to  
V
DDEXT or GND, except for the following:  
• DAI_Px, SPICLK, MISO, MOSI, EMU, TMS,TRST, TDI  
and AD15-0 (NOTE: These pins have internal pull-up  
resistors.)  
The following symbols appear in the Type column of Table 2:  
A = Asynchronous, G = Ground, I = Input, O = Output,  
P = Power Supply, S = Synchronous, (A/D) = Active Drive,  
(O/D) = Open Drain, and T = Three-State.  
Table 2. Pin Descriptions  
Pin  
Type  
State During &  
After Reset  
Function  
AD15-0  
I/O/T  
Three-state with  
pull-up enabled  
Parallel Port Address/Data. The ADSP-21267 parallel port and its corresponding  
DMA unit output addresses and data for peripherals on these multiplexed pins. The  
multiplex state is determined by the ALE pin. The parallel port can operate in either  
8-bit or 16-bit mode. Each AD pin has a 22.5 Kinternal pull-up resistor. See Address  
Data Modes on page 13 for details of the AD pin operation:  
For 8-bit mode: ALE is automatically asserted whenever a change occurs in the  
upper 16 external address bits, A23-8; ALE is used in conjunction with an external  
latch to retain the values of the A23-8.  
For 16-bit mode: ALE is automatically asserted whenever a change occurs in the  
address bits, A15-0; ALE is used in conjunction with an external latch to retain the  
values of the A15-0.  
To use these pins as flags (FLAG15-0) set (=1) bit 20 of the SYSCTL register and  
disable the parallel port. See Table 3 on page 13 for a list of how the AD15-0 pins  
map to the flag pins. When used as an input, the IDP Channel0 can use these pins  
for parallel input data.  
RD  
O
O
O
Output only, driven Parallel Port Read Enable. RD is asserted low whenever the DSP reads 8-bit or 16-  
high1  
bit data from an external memory device. When AD15-0 are flags, this pin remains  
deasserted.  
WR  
ALE  
Output only, driven Parallel Port Write Enable. WR is asserted low whenever the DSP writes 8-bit or  
high1  
16-bit data to an external memory device. When AD15-0 are flags, this pin remains  
deasserted.  
Output only, driven Parallel Port Address Latch enable. ALE is asserted whenever the DSP drives a  
low1  
new address on the parallel port address pins. On reset, ALE is active high. However,  
it can be reconfigured using software to be active low. When AD15-0 are flags, this  
pin remains deasserted.  
FLAG3-0  
I/O/A  
Three-state  
Flag Pins. Each FLAG pin is configured via control bits as either an input or output.  
As an input, it can be tested as a condition. As an output, it can be used to signal  
external peripherals. These pins can be used as an SPI interface slave select output  
during SPI mastering. These pins are also multiplexed with the IRQx and the TIMEXP  
signals.  
In SPI master boot mode, FLAG0 is the slave select pin that must be connected to  
an SPI EPROM. FLAG0 is configured as a slave select during SPI master boot. When  
bit 16 is set (=1) in the SYSCTL register, FLAG0 is configured as IRQ0.  
When bit 17 is set (=1) in the SYSCTL register, FLAG1 is configured as IRQ1.  
When bit 18 is set (=1) in the SYSCTL register, FLAG2 is configured as IRQ2.  
When bit 19 is set (=1) in the SYSCTL register, FLAG3 is configured as TIMEXP which  
indicates that the system timer has expired.  
Rev. PrA  
|
Page 10 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Table 2. Pin Descriptions (Continued)  
Pin  
Type  
State During &  
After Reset  
Function  
Digital Audio Interface Pins. These pins provide the physical interface to the SRU.  
DAI_P20-1  
I/O/T  
Three-state with  
programmable pull- The SRU configuration registers define the combination of on-chip peripheral  
up  
inputs or outputs connected to the pin and to the pin’s output enable. The config-  
uration registers of these peripherals then determines the exact behavior of the pin.  
Any input or output signal present in the SRU may be routed to any of these pins.  
The SRU provides the connection from the Serial ports, Input data port, precision  
clock generators and timers to the DAI_P20-1 pins These pins have internal 22.5 KΩ  
pull-up resistors which are enabled on reset. These pull-ups can be disabled in the  
DAI_PIN_PULLUP register.  
SPICLK  
I/O  
Three-state with  
pull-up enabled  
Serial Peripheral Interface Clock Signal. Driven by the master, this signal controls  
the rate at which data is transferred. The master may transmit data at a variety of  
baud rates. SPICLK cycles once for each bit transmitted. SPICLK is a gated clock that  
is active during data transfers, only for the length of the transferred word. Slave  
devices ignore the serial clock if the slave select input is driven inactive (HIGH).  
SPICLK is used to shift out and shift in the data driven on the MISO and MOSI lines.  
The data is always shifted out on one clock edge and sampled on the opposite edge  
of the clock. Clock polarity and clock phase relative to data are programmable into  
the SPICTL control register and define the transfer format. SPICLK has a 22.5 KΩ  
internal pull-up resistor.  
SPIDS  
I
Input only  
SerialPeripheralInterfaceSlaveDeviceSelect. Anactivelowsignalusedtoselect  
the DSP as an SPI slave device. This input signal behaves like a chip select, and is  
provided by the master device for the slave devices. In multi-master mode the DSPs  
SPIDS signal can be driven by a slave device to signal to the DSP (as SPI master) that  
an error has occurred, as some other device is also trying to be the master device.  
If asserted low when the device is in master mode, it is considered a multi-master  
error. For asingle-master, multiple-slave configurationwhereflag pins areused, this  
pin must be tied or pulled high to VDDEXT on the master device. For ADSP-21267 to  
ADSP-21267 SPI interaction, any of the master ADSP-21267's flag pins can be used  
to drive the SPIDS signal on the ADSP-21267 SPI slave device.  
MOSI  
MISO  
I/O (O/D)  
I/O (O/D)  
Three-state with  
pull-up enabled  
SPI Master Out Slave In. If the ADSP-21267 is configured as a master, the MOSI pin  
becomes a data transmit (output) pin, transmitting output data. If the ADSP-21267  
is configured as a slave, the MOSI pin becomes a data receive (input) pin, receiving  
input data. In an ADSP-21267 SPI interconnection, the data is shifted out from the  
MOSI output pin of the master and shifted into the MOSI input(s) of the slave(s).  
MOSI has a 22.5 Kinternal pull-up resistor.  
Three-state with  
pull-up enabled  
SPI Master In Slave Out. If the ADSP-21267 is configured as a master, the MISO pin  
becomes a data receive (input) pin, receiving input data. If the ADSP-21267 is  
configured as a slave, the MISO pin becomes a data transmit (output) pin, trans-  
mitting output data. In an ADSP-21267 SPI interconnection, the data is shifted out  
from the MISO output pin of the slave and shifted into the MISO input pin of the  
master. MISO has a 22.5Kinternal pull-up resistor. MISO can be configured as O/D  
by setting the OPD bit in the SPICTL register.  
Note: Only one slave is allowed to transmit data at any given time. To enable  
broadcast transmission to multiple SPI-slaves, the DSP's MISO pin may be disabled  
by setting (=1) bit 5 (DMISO) of the SPICTL register.  
BOOTCFG1-0  
I
Input only  
Boot Configuration Select. Selects the boot mode for the DSP. The BOOTCFG pins  
must be valid before reset is asserted. See Table 4 on page 13 for a description of  
the boot modes.  
Rev. PrA  
|
Page 11 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Table 2. Pin Descriptions (Continued)  
Pin  
Type  
State During &  
After Reset  
Function  
CLKIN  
I
Input only  
LocalClockIn. UsedinconjunctionwithXTAL. CLKINistheADSP-21267clockinput.  
It configures theADSP-21267 to use either its internal clock generator or an external  
clock source. Connecting the necessary components to CLKIN and XTAL enables  
the internal clock generator. Connecting the external clock to CLKIN while leaving  
XTAL unconnected configures the ADSP-21267 to use the external clock source  
such as an external clock oscillator. The core is clocked either by the PLL output or  
this clockinput depending on theCLKCFG1-0pin settings. CLKIN maynotbe halted,  
changed, or operated below the specified frequency.  
XTAL  
O
I
Output only2  
Input only  
Crystal Oscillator Terminal. Used in conjunction with CLKIN to drive an external  
crystal.  
CLKCFG1-0  
Core/CLKIN Ratio Control. These pins set the start up clock frequency. See Table 5  
on page 13 for a description of the clock configuration modes.  
Note that the operating frequency can be changed by programming the PLL multi-  
plier and divider in the PMCTL register at any time after the core comes out of reset.  
RSTOUT/CLKOUT  
RESET  
O
Output only  
Input only  
Input only3  
Reset Out/Local Clock Out. Drives out the core reset signal to an external device.  
CLKOUT can also be configured as a reset out pin (RSTOUT). The functionality can  
be switched between the PLL output clock and reset out by setting bit 12 of the  
PMCTL register. The default is reset out.  
I/A  
Processor Reset. Resets the ADSP-21267 to a known state. Upon deassertion, there  
is a 4096 CLKIN cycle latency for the PLL to lock. After this time, the core begins  
program execution from the hardware reset vector address. The RESET input must  
be asserted (low) at power-up.  
TCK  
TMS  
TDI  
I
Test Clock (JTAG). Provides a clock for JTAG boundary scan. must be asserted  
(pulsed low) after power-up or held low for proper operation of the ADSP-21267.  
I/S  
I/S  
Three-state with  
pull-up enabled  
Test Mode Select (JTAG). Used to control the test state machine. TMS has a  
22.5 kinternal pull-up resistor.  
Three-state with  
pull-up enabled  
Test Data Input (JTAG). Provides serial data for the boundary scan logic. TDI has a  
22.5 kinternal pull-up resistor.  
TDO  
TRST  
O
Three-state4  
Test Data Output (JTAG). Serial scan output of the boundary scan path.  
I/A  
Three-state with  
pull-up enabled  
Test Reset (JTAG). Resets the test state machine. TRST must be asserted (pulsed  
low) after power-up or held low for proper operation of the ADSP-21267. TRST has  
a 22.5 kinternal pull-up resistor.  
EMU  
O (O/D)  
Three-state with  
pull-up enabled  
Emulation Status. Must be connected to the ADSP-21267 Analog Devices DSP  
Tools product line of JTAG emulators target board connector only. EMU has a  
22.5 kinternal pullup resistor.  
VDDINT  
VDDEXT  
AVDD  
P
P
P
Core Power Supply. Nominally +1.2 V dc and supplies the DSP’s core processor (13  
pins on the BGA package, 32 pins on the LQFP package).  
I/O Power Supply. Nominally +3.3 V dc. (6 pins on the BGA package, 10 pins on the  
LQFP package).  
Analog Power Supply. Nominally +1.2 V dc and supplies the DSP’s internal PLL  
(clock generator). This pin has the same specifications as VDDINT, except that added  
filtering circuitry is required. For more information, see Power Supplies on page 7.  
AVSS  
G
G
Analog Power Supply Return.  
GND  
Power Supply Return. (54 pins on the BGA package, 39 pins on the LQFP package).  
1 RD, WR, and ALE are continuously driven by the DSP and won’t be three-stated.  
2 Output only is a three-state driver with its output path always enabled.  
3 Input only is three-state driver with both output path.  
4 Three-state is three-state driver.  
Rev. PrA  
|
Page 12 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
deasserted. For 16-bit data transfers, ALE latches address bits  
A15-A0 when asserted, followed by data bits D15-D0 when  
deasserted.  
ADDRESS DATA PINS AS FLAGS  
To use these pins as flags (FLAG15-0) set (=1) bit 20 of the  
SYSCTL register and disable the parallel port.  
Table 6. Address/ Data Mode Selection  
Table 3. AD[15:0] to FLAG Pin Mapping  
EP Data  
Mode  
ALE  
AD7-0  
Function  
AD15-8  
Function  
AD Pin  
AD0  
Flag Pin  
FLAG8  
FLAG9  
FLAG10  
FLAG11  
FLAG12  
FLAG13  
FLAG14  
FLAG15  
FLAG0  
FLAG1  
FLAG2  
FLAG3  
FLAG4  
FLAG5  
FLAG6  
FLAG7  
8-bit  
Asserted  
A15-8  
D7-0  
A7-0  
D7-0  
A23-16  
A7-0  
AD1  
8-bit  
Deasserted  
Asserted  
AD2  
16-bit  
16-bit  
A15-8  
D15-8  
AD3  
Deasserted  
AD4  
AD5  
AD6  
AD7  
AD8  
AD9  
AD10  
AD11  
AD12  
AD13  
AD14  
AD15  
Boot Modes  
Table 4. Boot Mode Selection  
BOOTCFG1-0  
Booting Mode  
00  
01  
10  
11  
SPI Slave Boot  
SPI Master Boot  
Parallel Port boot via EPROM  
Internal Boot Mode (ROM code only)  
CORE INSTRUCTION RATE TO CLKIN RATIO MODES  
Table 5. Core Instruction Rate/ CLKIN Ratio Selection  
CLKCFG1-0  
Core to CLKIN Ratio  
00  
01  
10  
11  
3:1  
16:1  
8:1  
Reserved  
ADDRESS DATA MODES  
The following table shows the functionality of the AD pins for  
8-bit and 16-bit transfers to the parallel port. For 8-bit data  
transfers, ALE latches address bits A23-A8 when asserted, fol-  
lowed by address bits A7-A0 and data bits D7-D0 when  
Rev. PrA  
|
Page 13 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
ADSP-21267 SPECIFICATIONS  
RECOMMENDED OPERATING CONDITIONS  
K Grade  
Parameter1  
Min  
Max  
Unit  
VDDINT  
AVDD  
VDDEXT  
VIH  
Internal (Core) Supply Voltage  
1.14  
1.14  
3.13  
2.0  
1.26  
V
Analog (PLL) Supply Voltage  
1.26  
V
External (I/O) Supply Voltage  
3.47  
V
High Level Input Voltage2, @ VDDEXT = max  
Low Level Input Voltage2 @ VDDEXT = min  
Ambient Operating Temperature3 4  
VDDEXT+0.5  
0.8  
V
VIL  
-0.5  
0
V
TAMB  
+70  
°C  
1
Specifications subject to change without notice.  
2 Applies to input and bidirectional pins: AD15-0, FLAG3-0, DAI_Px, SPICLK, MOSI, MISO, SPIDS, BOOTCFGx, CLKIN, CLKCFGx, RESET, TCK, TMS, TDI, TRST.  
3 See Thermal Characteristics on page 37 for information on thermal specifications.  
4 See Engineer-to-Engineer Note (No. 216) for further information.  
ELECTRICAL CHARACTERISTICS  
Parameter1  
VOH  
Test Conditions  
Min  
Max  
Unit  
V
High Level Output Voltage2  
Low Level Output Voltage2  
High Level Input Current4, 5  
Low Level Input Current4  
Low Level Input Current Pull-Up5  
Three-State Leakage Current 6, 7, 8  
Three-State Leakage Current6  
Three-State Leakage Current Pull-Up7  
Supply Current (Internal)9, 10, 11  
Supply Current (Analog)12  
@ VDDEXT = min, IOH = -1.0 mA3  
@ VDDEXT = min, IOL = 1.0 mA3  
@ VDDEXT = max, VIN = VDDEXT max  
@ VDDEXT = max, VIN = 0 V  
@ VDDEXT = max, VIN = 0 V  
@ VDDEXT= max, VIN = VDDEXT max  
@ VDDEXT = max, VIN = 0 V  
@ VDDEXT = max, VIN = 0 V  
tCCLK = 5.0 ns, VDDINT = 1.2V, TAMB = +25°C  
AVDD = max  
2.4  
VOL  
0.4  
10  
V
IIH  
µA  
µA  
µA  
µA  
µA  
µA  
mA  
mA  
pF  
IIL  
10  
IILPU  
200  
10  
IOZH  
IOZL  
10  
IOZLPU  
IDD-INTYP  
AIDD  
CIN  
200  
500  
10  
Input Capacitance13, 14  
fIN=1 MHz, TCASE=25°C, VIN=1.2V  
4.7  
1 Specifications subject to change without notice.  
2 Applies to output and bidirectional pins: AD15-0, RD, WR, ALE, FLAG3-0, DAI_Px, SPICLK, MOSI, MISO, EMU, TDO, CLKOUT, XTAL.  
3 See Output Drive Currents on page 36 for typical drive current capabilities.  
4 Applies to input pins: SPIDS, BOOTCFGx, CLKCFGx, TCK, RESET, CLKIN.  
5 Applies to input pins with 22.5 Kinternal pull-ups: TRST, TMS, TDI.  
6 Applies to three-statable pins: FLAG3-0.  
7 Applies to three-statable pins with 22.5 kKpull-ups: AD15-0, DAI_Px, SPICLK, MISO, MOSI.  
8 Applies to open-drain output pins: EMU, MISO, MOSI.  
9 Typical internal current data reflects nominal operating conditions.  
10See Engineer-to-Engineer Note (No. 216) for further information.  
11Characterized, but not tested.  
12Characterized, but not tested.  
13Applies to all signal pins.  
14Guaranteed, but not tested.  
Rev. PrA  
|
Page 14 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
ABSOLUTE MAXIMUM RATINGS  
1
Internal (Core) Supply Voltage (VDDINT  
)
-0.3 V to +1.4 V  
-0.3 V to +1.4 V  
-0.3 V to +3.8 V  
-0.5 V to VDDEXT1 + 0.5 V  
-0.5 V to VDDEXT1 + 0.5 V  
200 pF  
1
Analog (PLL) Supply Voltage (AVDD  
)
1
External (I/O) Supply Voltage (VDDEXT  
Input Voltage  
)
Output Voltage Swing  
Load Capacitance1  
Storage Temperature Range1  
-65°C to +150°C  
125°C  
Junction Temperature under Bias  
1 Stresses greater than those listed above may cause permanent damage to the device. These are stress ratings  
only, and functional operation of the device at these or any other conditions greater than those indicated  
in the operational sections of this specification is not implied. Exposure to absolute maximum rating  
conditions for extended periods may affect device reliability.  
ESD SENSITIVITY  
CAUTION  
ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000V readily  
accumulate on the human body and test equipment and can discharge without detection.  
Although the ADSP-21267 features proprietary ESD protection circuitry, permanent  
damage may occur on devices subjected to high-energy electrostatic discharges.  
Therefore, proper ESD precautions are recommended to avoid performance degradation  
or loss of functionality.  
The ADSP-21267’s internal clock switches at higher frequencies  
than the system input clock (CLKIN). To generate the internal  
clock, the DSP uses an internal phase-locked loop (PLL). This  
PLL-based clocking minimizes the skew between the system  
clock (CLKIN) signal and the DSP’s internal clock (the clock  
source for the parallel port logic and I/O pads).  
TIMING SPECIFICATIONS  
The ADSP-21267’s internal clock (a multiple of CLKIN) pro-  
vides the clock signal for timing internal memory, processor  
core, serial ports, and parallel port (as required for read/write  
strobes in asynchronous access mode). During reset, program  
the ratio between the DSP’s internal clock frequency and exter-  
nal (CLKIN) clock frequency with the CLKCFG1-0 pins. To  
determine switching frequencies for the serial ports, divide  
down the internal clock, using the programmable divider con-  
trol of each port (DIVx for the serial ports).  
Note the definitions of various clock periods that are a function  
of CLKIN and the appropriate ratio control (Table 7).  
Table 7. ADSP-21267 CLKOUT and CCLK Clock Generation Operation  
Timing Requirements  
Description  
Input Clock  
Core Clock  
Calculation  
1/tCK  
CLKIN  
CCLK  
1/tCCLK  
Timing Requirements  
Description1  
tCK  
CLKIN Clock Period  
tCCLK  
(Processor) Core Clock Period  
Serial Port Clock Period = (tCCLK) x SR  
SPI Clock Period = (tCCLK) x SPIR  
tSCLK  
tSPICLK  
1 where:  
SR = serial port-to-core clock ratio (wide range, determined by SPORT CLKDIV)  
SPIR = SPI-to-Core Clock Ratio (wide range, determined by SPIBAUD register)  
DAI_Px = Serial Port Clock  
SPICLK = SPI Clock  
Rev. PrA  
|
Page 15 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Figure 5 shows Core to CLKIN ratios of 3:1, 8:1 and 16:1 with  
external oscillator or crystal. Note that more ratios are possible  
and can be set through software using the power management  
control register (PMCTL). For more information, see the ADSP-  
2126x SHARC DSP Core Manual.  
CLKOUT  
CLKIN  
XTAL  
OSC  
PLL  
3:1, 8:1,  
16:1  
CCLK  
(CORE CLOCK)  
PLLILCLK  
XTAL  
CLK-CFG [1:0]  
Figure 5. Core Clock and System Clock Relationship to CLKIN  
Use the exact timing information given. Do not attempt to  
derive parameters from the addition or subtraction of others.  
While addition or subtraction would yield meaningful results  
for an individual device, the values given in this data sheet  
reflect statistical variations and worst cases. Consequently, it is  
not meaningful to add parameters to derive longer times.  
See Figure 30 on page 36 under Test Conditions for voltage ref-  
erence levels.  
Switching Characteristics specify how the processor changes its  
signals. Circuitry external to the processor must be designed for  
compatibility with these signal characteristics. Switching char-  
acteristics describe what the processor will do in a given  
circumstance. Use switching characteristics to ensure that any  
timing requirement of a device connected to the processor (such  
as memory) is satisfied.  
Timing Requirements apply to signals that are controlled by cir-  
cuitry external to the processor, such as the data input for a read  
operation. Timing requirements guarantee that the processor  
operates correctly with other devices.  
The ADSP-21267’s internal clock (a multiple of CLKIN) pro-  
vides the clock signal for timing internal memory, processor  
core, serial ports, and parallel port (as required for read/write  
strobes in asynchronous access mode). During reset, program  
the ratio between the DSP’s internal clock frequency and exter-  
nal (CLKIN) clock frequency with the CLKCFG1-0 pins. To  
determine switching frequencies for the serial ports, divide  
down the internal clock, using the programmable divider con-  
trol of each port (DIVx for the serial ports).  
The ADSP-21267’s internal clock switches at higher frequencies  
than the system input clock (CLKIN). To generate the internal  
clock, the DSP uses an internal phase-locked loop (PLL). This  
PLL-based clocking minimizes the skew between the system  
clock (CLKIN) signal and the DSP’s internal clock (the clock  
source for the parallel port logic and I/O pads).  
Note the following definitions of various clock periods that are a  
function of CLKIN and the appropriate ratio control.  
Rev. PrA  
|
Page 16 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Power up Sequencing  
The timing requirements for DSP startup are given in Table 8.  
Table 8. Power Up Sequencing Timing Requirements (DSP Startup)  
Parameter  
Min  
Max  
Unit  
Timing Requirements  
tRSTVDD  
tIVDDEVDD  
tCLKVDD  
tCLKRST  
tPLLRST  
tWRST  
RESET low before VDDINT/VDDEXT on  
VDDINT on before VDDEXT  
CLKIN valid after VDDINT/VDDEXT valid1  
CLKIN valid before RESET deasserted  
PLL control setup before RESET deasserted  
Subsequent RESET low pulse width4  
0
ns  
ms  
ms  
µs  
µs  
ns  
-50  
0
102  
203  
4tCK  
200  
200  
Switching Characteristic  
4,5  
tCORERST  
DSP core reset deasserted after RESET deasserted  
4096tCK  
1 Valid VDDINT/VDDEXT assumes that the supplies are fully ramped to their 1.2 and 3.3 volt rails. Voltage ramp rates can vary from microseconds to hundreds of milliseconds  
depending on the design of the power supply subsystem.  
2 Assumes a stable CLKIN signal, after meeting worst-case startup timing of crystal oscillators. Refer to your crystal oscillator manufacturer's datasheet for startup time.  
Assume a 25 ms maximum oscillator startup time if using the XTAL pin and internal oscillator circuit in conjunction with an external crystal.  
3 Based on CLKIN cycles  
4 Applies after the power-up sequence is complete. Subsequent resets require a minimum of 4 CLKIN cycles for RESET to be held low in order to properly initialize and  
propagate default states at all I/O pins.  
5 The 4096 cycle count depends on tSRST specification in Table 10. If setup time is not met, 1 additional CLKIN cycle may be added to the core reset time, resulting in 4097  
cycles maximum.  
RESET  
tRSTVDD  
V
DDINT  
tIVDDEVDD  
V
tCLKVDD  
DDEXT  
CLKIN  
tCLKRST  
CLK_CFG1-0  
tCO RERST  
tPLLRST  
RSTOUT*  
*MULTIPLEXED W ITH CLKOUT  
Figure 6. Power Up Sequencing  
Rev. PrA  
|
Page 17 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Clock Input  
Table 9. Clock Input  
Parameter  
150 MHz  
Min  
200 MHz  
Min  
Unit  
Max  
Max  
Timing Requirements  
tCK  
CLKIN Period  
201  
7.51  
7.51  
1602  
802  
802  
3
151  
61  
61  
1602  
802  
802  
3
ns  
ns  
ns  
ns  
ns  
tCKL  
tCKH  
tCKRF  
tCCLK  
CLKIN Width Low  
CLKIN Width High  
CLKIN Rise/Fall (0.4V-2.0V)  
CCLK Period3  
6.66  
10  
5
10  
1 Applies only for CLKCFG1-0 = 00 and default values for PLL control bits in PMCTL.  
2 Applies only for CLKCFG1-0 = 01 and default values for PLL control bits in PMCTL.  
3 Any changes to PLL control bits in the PMCTL register must meet core clock timing specification tCCLK  
.
CLKIN  
XTAL  
1M  
tCK  
CLKIN  
tCKH  
tCKL  
C1  
C2  
X1  
Figure 7. Clock Input  
NOTE: C1 AND C2 ARE SPECIFIC TO CRYSTAL SPECIFIED FOR X1.  
CONTACT CRYSTAL MANUFACTURER FOR DETAILS. CRYSTAL  
SELECTION MUST COMPLY WITH CLKCFG1-0 = 10 OR = 01.  
Clock Signals  
The ADSP-21267 can use an external clock or a crystal. See  
CLKIN pin description. The programmer can configure the  
ADSP-21267 to use its internal clock generator by connecting  
the necessary components to CLKIN and XTAL. Figure 8 shows  
the component connections used for a crystal operating in fun-  
Figure 8. 150 MHz or 200 MHz Operation (Fundamental Mode Crystal)  
Reset  
Table 10. Reset  
Parameter  
Min  
Max  
Unit  
Timing Requirements  
tWRST  
tSRST  
RESET Pulse Width Low1  
4tCK  
8
ns  
ns  
RESET Setup Before CLKIN Low  
1 Applies after the power-up sequence is complete. At power-up, the processor's internal phase-locked loop requires no more than 100 µs while RESET is low, assuming  
stable VDD and CLKIN (not including start-up time of external clock oscillator).  
CLKIN  
tWRST  
tSRST  
RESET  
Figure 9. Reset  
Rev. PrA  
|
Page 18 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Interrupts  
The following timing specification applies to the FLAG0,  
FLAG1, and FLAG2 pins when they are configured as IRQ0,  
IRQ1, and IRQ2 interrupts. Also applies to DAI_P[20:1] pins  
when configured as interrupts.  
Table 11. Interrupts  
Parameter  
Min  
Max  
Unit  
Timing Requirement  
tIPW  
IRQx Pulse Width  
2 x tCCLK +2  
ns  
DAI_P[20:1]  
(FLG2-0)  
(IRQ2-0)  
tIPW  
Figure 10. Interrupts  
Core Timer  
The following timing specification applies to FLAG3 when it is  
configured as the core timer (CTIMER).  
Table 12. Core Timer  
Parameter  
Min  
Max  
Unit  
Switching Characteristic  
tWCTIM  
CTIMER Pulse Width  
4 x tCCLK  
ns  
FLG3  
(CTIM ER)  
tW CTIM  
Figure 11. Core Timer  
Rev. PrA  
|
Page 19 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Timer PWM_OUT Cycle Timing  
The following timing specification applies to Timer[2:0] in  
PWM_OUT (pulse width modulation) mode. Timer signals are  
routed to the DAI_P[20:1] pins through the SRU. Therefore, the  
timing specifications provided below are valid at the  
DAI_P[20:1] pins.  
Table 13. Timer[2:0] PWM_OUT Timing  
Parameter  
Min  
2 tCCLK  
Max  
Unit  
Switching Characteristic  
tPWMO  
Timer[2:0] Pulse Width Output  
2(231 – 1) tCCLK  
ns  
tPWMO  
DAI_P[20:1]  
(TIMER[2:0])  
Figure 12. Timer[2:0] PWM_OUT Timing  
Timer WDTH_CAP Timing  
The following timing specification applies to Timer[2:0] in  
WDTH_CAP (pulse width count and capture) mode. Timer sig-  
nals are routed to the DAI_P[20:1] pins through the SRU.  
Therefore, the timing specifications provided below are valid at  
the DAI_P[20:1] pins.  
Table 14. Timer[2:0] Width Capture Timing  
Parameter  
Min  
2 tCCLK  
Max  
Unit  
Timing Requirement  
tPWI  
Timer[2:0] Pulse Width  
2(231 – 1) tCCLK  
ns  
tPWI  
DAI_P[20:1]  
(TIMER[2:0])  
Figure 13. Timer[2:0] Width Capture Timing  
Rev. PrA  
|
Page 20 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
DAI Pin to Pin Direct Routing  
For direct pin connections only (for example DAI_PB01_I to  
DAI_PB02_O).  
Table 15. DAI Pin to Pin Routing  
Parameter  
Min  
Max  
Unit  
Timing Requirement  
tDPIO  
Delay DAI Pin Input Valid to DAI Output Valid  
3
10  
ns  
DAI_Pn  
DAI_Pm  
tDPIO  
Figure 14. DAI Pin to PIN Direct Routing  
Rev. PrA  
|
Page 21 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
inputs and outputs are not directly routed to/from DAI pins (via  
pin buffers) there is not timing data available. All Timing  
Parameters and Switching Characteristics apply to external DAI  
pins (DAI_P07 – DAI_P20).  
Precision Clock Generator (Direct Pin Routing)  
This timing is only valid when the SRU is configured such that  
the Precision Clock Generator (PCG) takes its inputs directly  
from the DAI pins (via pin buffers) and sends its outputs  
directly to the DAI pins. For the other cases, where the PCG’s  
Table 16. Precision Clock Generator (Direct Pin Routing)  
Parameter  
Min  
Max  
Unit  
Timing Requirements  
tPCGIW  
tSTRIG  
tHTRIG  
Input Clock Pulse Width  
20  
2
ns  
ns  
ns  
PCG Trigger Setup Before Falling Edge of PCG Input Clock  
PCG Trigger Hold After Falling Edge of PCG Input Clock  
2
Switching Characteristics  
tDPCGIO  
PCG Output Clock and Frame Sync Active Edge Delay After PCG Input  
Clock Falling Edge  
2.5  
10  
ns  
tDTRIG  
PCG Output Clock and Frame Sync Delay After PCG Trigger  
Output Clock Pulse Width  
2.5 + 2.5 x tPCGOW 10 + 2.5 x tPCGOW ns  
40 ns  
tPCGOW  
tSTRIG  
DAI_Pn  
PCG_TRIGx_I  
tHTRIG  
tPCGIW  
DAI_Pm  
PCG_EXTx_I  
(CLKIN)  
tDPCGIO  
DAI_Py  
PCG_CLKx_O  
tPCGOW  
DAI_Pz  
PCG_FSx_O  
tDTRIG  
Figure 15. Precision Clock Generator (Direct Pin Routing)  
Rev. PrA  
|
Page 22 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Flags  
The timing specifications provided below apply to the  
FLAG[3:0] and DAI_P[20:1] pins, the parallel port and the  
serial peripheral interface (SPI). See Table 2, “Pin Descriptions,”  
on page 10 for more information on flag use.  
Table 17. Flags  
Parameter  
Min  
Max  
Unit  
Timing Requirement  
tFIPW  
FLAG[3:0] IN Pulse Width  
2 x tCCLK+3  
ns  
Switching Characteristic  
tFOPW FLAG[3:0] OUT Pulse Width  
2 x tCCLK – 1  
ns  
DAI_P[20:1]  
(FLG3-0IN  
)
(AD[15:0])  
tFIPW  
DAI_P[20:1]  
(FLG3-0OUT  
)
(AD[15:0])  
tFOPW  
Figure 16. Flags  
Rev. PrA  
|
Page 23 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Memory Read–Parallel Port  
Use these specifications for asynchronous interfacing to memo-  
ries (and memory-mapped peripherals) when the ADSP-21267  
is accessing external memory space.  
Table 18. 8-Bit Memory Read Cycle  
Parameter  
Min  
Max  
Unit  
Timing Requirements  
tDRS  
tDRH  
tDAD  
Address/Data [7:0] Setup Before RD High  
Address/Data [7:0] Hold After RD High  
Address [15:8] to Data Valid  
3.3  
0
ns  
ns  
ns  
D + 0.5 x tCCLK – 3.5  
Switching Characteristics  
tALEW  
tALERW  
tADAS  
tADAH  
tALEHZ  
tRW  
ALE Pulse Width  
2 x tCCLK – 2  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ALE Deasserted to Read/Write Asserted  
Address/Data [15:0] Setup Before ALE Deasserted1  
Address/Data [15:0] Hold After ALE Deasserted1  
ALE Deasserted1 to Address/Data[7:0] In High Z  
RD Pulse Width  
1 x tCCLK – 1  
2.5 x tCCLK – 2.0  
0.5 x tCCLK – 0.8  
0.5 x tCCLK – 0.8  
D – 2  
0.5 x tCCLK + 3.0  
tADRH  
Address/Data [15:8] Hold After RD High  
0.5 x tCCLK – 1 + H  
D = (Data Cycle Duration) x tCCLK  
H= tCCLK (if a hold cycle is specified, else H = 0)  
1 On reset, ALE is an active high cycle. However, it can be reconfigured by software to be active low.  
tALEW  
ALE  
tALERW  
tRW  
RD  
WR  
tALEHZ  
tADAH  
tADAS  
tADRH  
AD[15:8]  
VALID ADDRESS  
VALID ADDRESS  
VALID ADDRESS  
tDRS  
tDRH  
AD[7:0]  
VALID DATA  
tDAD  
Figure 17. Read Cycle For 8-bit Memory Timing  
Rev. PrA  
|
Page 24 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Table 19. 16-bit Memory Read Cycle  
Parameter  
Min  
Max  
Unit  
Timing Requirements  
tDRS  
tDRH  
Address/Data [15:0] Setup Before RD high  
Address/Data [15:0] Hold After RD high  
3.3  
0
ns  
ns  
Switching Characteristics  
ns  
ns  
ns  
ns  
ns  
tALEW  
tALERW  
tADAS  
tADAH  
tALEHZ  
tRW  
ALE Pulse Width  
2 x tCCLK – 2  
1 x tCCLK – 1  
2.5 x tCCLK – 2.0  
0.5 x tCCLK – 0.8  
0.5 x tCCLK – 0.8  
D – 2  
ALE Deasserted to Read/Write Asserted  
Address/Data [15:0] Setup Before ALE Deasserted1  
Address/Data [15:0] Hold After ALE Deaserted1  
ALE Deasserted1 to Address/Data[15:0] In High Z  
RD Pulse Width  
0.5tCCLK + 3.0  
ns  
ns  
D = (Data Cycle Duration) x tCCLK  
H = tCCLK (if a hold cycle is specified, else H = 0)  
1 On reset, ALE is an active high cycle. However, it can be reconfigured by software to be active low.  
ALE  
tALEW  
tALERW  
RD  
tRW  
WR  
tADAH  
tDRS  
tDRH  
tADAS  
VALID DATA  
AD[15:0]  
VALID ADDRESS  
tALEHZ  
Figure 18. Read Cycle For 16-bit Memory Timing  
Rev. PrA  
|
Page 25 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Memory Write—Parallel Port  
Use these specifications for asynchronous interfacing to memo-  
ries (and memory-mapped peripherals) when the  
ADSP-21267 is accessing external memory space.  
Table 20. 8-bit Memory Write Cycle  
Parameter  
Min  
Max  
Unit  
Switching Characteristics:  
tALEW  
tALERW  
tADAS  
tADAH  
tWW  
ALE Pulse Width  
2 x tCCLK – 2  
1 x tCCLK – 1  
2.5 x tCCLK – 2.0  
0.5 x tCCLK – 0.8  
D - 2  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ALE Deasserted to Read/Write Asserted  
Address/Data [15:0] Setup Before ALE Deasserted1  
Address/data [15:0] Hold After ALE Deasserted1  
WR Pulse Width  
tADWL  
tADWH  
tALEHZ  
tDWS  
Address/Data [15:8] to WR Low  
0.5 x tCCLK – 1.5  
0.5 x tCCLK – 1 + H  
0.5 x tCCLK – 0.8  
D
Address/Data [15:8] hold after WR High  
ALE Deasserted1 to Address/Data[15:0] In High Z  
Address/Data [7:0] Setup Before WR High  
Address/Data [7:0] Hold After WR High  
Address/Data to WR High  
0.5tCCLK + 3.0 ns  
ns  
ns  
ns  
tDWH  
0.5 x tCCLK – 1.5 + H  
D
tDAWH  
D = (Data Cycle Duration) x tCCLK  
H = tCCLK (if a hold cycle is specified, else H = 0)  
1 On reset, ALE is an active high cycle. However, it can be reconfigured by software to be active low.  
tALERW  
ALE  
tALEW  
tDAWH  
WR  
tWW  
RD  
tALEHZ  
tADAH  
tADWL  
tADWH  
tADAS  
AD[15:8]  
VALID ADDRESS  
VALID ADDRESS  
tDWS  
tDWH  
VALID ADDRESS  
VALID DATA  
AD[7:0]  
Figure 19. Write Cycle For 8-bit Memory Timing  
Rev. PrA  
|
Page 26 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Table 21. 16-bit Memory Write Cycle  
Parameter  
Min  
Max  
Unit  
Switching Characteristics  
tALEW  
tALERW  
tADAS  
tADAH  
tWW  
ALE Pulse Width  
2 x tCCLK – 2  
1 x tCCLK – 1  
2.5 x tCCLK – 2.0  
0.5 x tCCLK – 0.8  
D – 2  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ALE Deasserted to Read/Write Asserted  
Address/Data [15:0] Setup Before ALE Deasserted1  
Address/Data [15:0] Hold After ALE Deasserted1  
WR Pulse Width  
ALE Deasserted1 to Address/Data[15:0] In High Z  
Address/Data [15:0] Setup Before WR High  
Address/Data [15:0] Hold After WR High  
tALEHZ  
tDWS  
0.5 x tCCLK – 0.8  
D
0.5tCCLK + 3.0  
tDWH  
0.5 x tCCLK – 1.5 + H  
D = (Data Cycle Duration) x tCCLK  
H = tCCLK (if a hold cycle is specified, else H = 0)  
1 On reset, ALE is an active high cycle. However, it can be reconfigured by software to be active low.  
ALE  
tALEW  
tALERW  
tWW  
WR  
RD  
tALEHZ  
tADAS  
tADAH  
tDWS  
tDWH  
VALID ADDRESS  
AD[15:0]  
VALID DATA  
Figure 20. Write Cycle For 16-bit Memory Timing  
Rev. PrA  
|
Page 27 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Serial port signals (SCLK, FS, DxA,/DxB) are routed to the  
Serial Ports  
DAI_P[20:1] pins using the SRU. Therefore, the timing specifi-  
cations provided below are valid at the DAI_P[20:1] pins.  
To determine whether communication is possible between two  
devices at clock speed n, the following specifications must be  
confirmed: 1) frame sync delay and frame sync setup and hold,  
2) data delay and data setup and hold, and 3) SCLK width.  
Table 22. Serial Ports—External Clock  
Parameter  
Min  
Max  
Unit  
Timing Requirements  
tSFSE  
FS Setup Before SCLK  
(Externally Generated FS in Either Transmit or Receive Mode)1  
2.5  
ns  
tHFSE  
FS Hold After SCLK  
(Externally Generated FS in Either Transmit or Receive Mode)1  
2.5  
2.5  
2.5  
7
ns  
ns  
ns  
ns  
ns  
tSDRE  
tHDRE  
tSCLKW  
tSCLK  
Receive Data Setup Before Receive SCLK1  
Receive Data Hold After SCLK1  
SCLK Width  
SCLK Period  
20  
Switching Characteristics  
tDFSE  
FS Delay After SCLK  
(Internally Generated FS in Ether Transmit or Receive Mode) 2  
7
7
ns  
tHOFSE  
FS Hold After SCLK  
(Internally Generated FS in Either Transmit or Receive Mode2)  
2
2
ns  
ns  
ns  
tDDTE  
tHDTE  
Transmit Data Delay After Transmit SCLK2  
Transmit Data Hold After Transmit SCLK2  
1 Referenced to sample edge.  
2 Referenced to drive edge.  
Table 23. Serial Ports—Internal Clock  
Parameter  
Min  
Max  
Unit  
Timing Requirements  
tSFSI  
FS Setup Before SCLK  
(Externally Generated FS in Either Transmit or Receive Mode)1  
6
ns  
tHFSI  
FS Hold After SCLK  
(Externally Generated FS in Either Transmit or Receive Mode)1  
1.5  
6
ns  
ns  
ns  
tSDRI  
tHDRI  
Receive Data Setup Before SCLK1  
Receive Data Hold After SCLK1  
1.5  
Switching Characteristics  
tDFSI  
FS Delay After SCLK (Internally Generated FS in Transmit Mode)2  
3
ns  
ns  
ns  
ns  
ns  
ns  
ns  
tHOFSI  
tDFSI  
tHOFSI  
tDDTI  
FS Hold After SCLK (Internally Generated FS in Transmit Mode)2  
FS Delay After SCLK (Internally Generated FS in Receive or Mode)2  
FS Hold After SCLK (Internally Generated FS in Receive Mode)2  
Transmit Data Delay After SCLK2  
-1.0  
-1.0  
3
3
tHDTI  
Transmit Data Hold After SCLK2  
-1.0  
tSCLKIW  
Transmit or Receive SCLK Width  
0.5tSCLK – 2  
0.5tSCLK + 2  
1 Referenced to the sample edge.  
2 Referenced to drive edge.  
Rev. PrA  
|
Page 28 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Table 24. Serial Ports—Enable and Three-State  
Parameter  
Min  
2
Max  
Unit  
Switching Characteristics  
tDDTEN  
tDDTTE  
tDDTIN  
Data Enable from External Transmit SCLK1  
Data Disable from External Transmit SCLK1  
Data Enable from Internal Transmit SCLK1  
ns  
ns  
ns  
7
-1  
1 Referenced to drive edge.  
Table 25. Serial Ports—External Late Frame Sync  
Parameter  
Min  
Max  
Unit  
Switching Characteristics  
tDDTLFSE  
Data Delay from Late External Transmit FS or External Receive FS  
with MCE = 1, MFD = 01  
7
ns  
ns  
tDDTENFS  
Data Enable for MCE = 1, MFD = 01  
0.5  
1 The tDDTLFSE and tDDTENFS parameters apply to Left-justified Sample Pair as well as DSP serial mode, and MCE = 1, MFD = 0.  
EXTERNAL RECEIVE FS WITH MCE = 1, MFD = 0  
DRIVE  
SAMPLE  
tHFSE/I  
DRIVE  
DIA_P[20:0]  
(SCLK)  
tSFSE/I  
DIA_P[20:0]  
(FS)  
tDDTE/I  
tDDTENFS  
tHDTE/I  
1ST BIT  
DIA_P[20:0]  
(D A/D B)  
X
X
2ND BIT  
tDDTLFSE  
LATE EXTERNAL TRANSMIT FS  
DRIVE  
SAMPLE  
tHFSE/I  
DRIVE  
DIA_P[20:0]  
(SCLK)  
tSFSE/I  
DIA_P[20:0]  
(FS)  
tDDTE/I  
tDDTENFS  
tHDTE/I  
1ST BIT  
DIA_P[20:0]  
(D A/D B)  
X
X
2ND BIT  
tDDTLFSE  
NOTE  
SERIAL PORT SIGNALS (SCLK, FS, DXA,/DXB) ARE ROUTED TO THE DAI_P[20:1] PINS USING THE SRU.  
THE TIMING SPECIFICATIONS PROVIDED HERE ARE VALID AT THE DAI_P[20:1] PINS.  
Figure 21. External Late Frame Sync1  
1 This figure reflects changes made to support Left-justified Sample Pair mode.  
Rev. PrA  
|
Page 29 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
DATA RECEIVE— INTERNAL CLOCK  
DRIVE EDGE SAMPLE EDGE  
DATA RECEIVE— EXTERNAL CLOCK  
DRIVE EDGE SAMPLE EDGE  
tSCLKIW  
tSCLKW  
DAI_P[20:1]  
(SCLK)  
DAI_P[20:1]  
(SCLK)  
tDFSI  
tDFSE  
tHFSE  
tHFSI  
tSFSI  
tSFSE  
tHOFSI  
tHOFSE  
DAI_P[20:1]  
(FS)  
DAI_P[20:1]  
(FS)  
tHDRE  
tSDRI  
tHDRI  
tSDRE  
DAI_P[20:1]  
DAI_P[20:1]  
(D A/D B)  
(D A/D B)  
X X  
X
X
NOTE: EITHER THE RISING EDGE OR FALLING EDGE OF SCLK (EXTERNAL), SCLK (INTERNAL) CAN BE USED AS THE ACTIVE SAMPLING EDGE.  
DATA TRANSMIT — INTERNAL CLOCK  
DRIVE EDGE SAMPLE EDGE  
DATA TRANSMIT — EXTERNAL CLOCK  
DRIVE EDGE SAMPLE EDGE  
tSCLKIW  
tSCLKW  
DAI_P[20:1]  
(SCLK)  
DAI_P[20:1]  
(SCLK)  
tDFSI  
tDFSE  
tHFSI  
tHOFSI  
tSFSI  
tHOFSE  
tSFSE  
tHFSE  
DAI_P[20:1]  
(FS)  
DAI_P[20:1]  
(FS)  
tDDTE  
tDDTI  
tHDTE  
tHDTI  
DAI_P[20:1]  
DAI_P[20:1]  
(D A/D B)  
(D A/D B)  
X X  
X
X
NOTE: EITHER THE RISING EDGE OR FALLING EDGE OF SCLK (EXTERNAL), SCLK (INTERNAL) CAN BE USED AS THE ACTIVE SAMPLING EDGE.  
DRIVE EDGE  
DRIVE EDGE  
DAI_P[20:1]  
SCLK (EXT)  
SCLK  
tDDTEN  
tDDTTE  
DAI_P[20:1]  
D
A/D B  
X
X
DRIVE EDGE  
DAI_P[20:1]  
SCLK (INT)  
tDDTIN  
DAI_P[20:1]  
D
A/D B  
X
X
Figure 22. Serial Ports  
Rev. PrA  
|
Page 30 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Input Data Port (IDP)  
The timing requirements for the IDP are given in Table 26. IDP  
Signals (SCLK, FS, SDATA) are routed to the DAI_P[20:1] pins  
using the SRU. Therefore, the timing specifications provided  
below are valid at the DAI_P[20:1] pins.  
Table 26. Input Data Port  
Parameter  
Min  
Max  
Unit  
Timing Requirements  
tSISFS  
FS Setup Before SCLK Rising Edge1  
FS Hold After SCLK Rising Edge1  
SData Setup Before SCLK Rising Edge1  
SData Hold After SCLK Rising Edge1  
Clock Width  
2.5  
2.5  
2.5  
2.5  
7
ns  
ns  
ns  
ns  
ns  
ns  
tSIHFS  
tSISD  
tSIHD  
tIDPCLKW  
tIDPCLK  
Clock Period  
20  
1
DATA, SCLK, FS can come from any of the DAI pins. SCLK and FS can also come via the Precision Clock Generators (PCG) or SPORTs. PCG's input can be either  
CLKIN or any of the DAI pins.  
SAMPLE EDGE  
tIDPCLKW  
DAI_P[20:1]  
(SCLK)  
tSIHFS  
tSISFS  
DAI_P[20:1]  
(FS)  
tSISD  
tSIHD  
DAI_P[20:1]  
(SDATA)  
Figure 23. IDP Master Timing  
Rev. PrA  
|
Page 31 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
that the most significant 16 bits of external PDAP data can be  
provided through either the parallel port AD[15:0] or the  
DAI_P[20:5] pins. The remaining 4 bits can only be sourced  
through DAI_P[4:1]. The timing below is valid at the  
DAI_P[20:1] pins or at the AD[15:0] pins.  
Parallel Data Acquisition Port (PDAP)  
The timing requirements for the PDAP are provided in  
Table 27. PDAP is the parallel mode operation of channel 0 of  
the IDP. For details on the operation of the IDP, see the IDP  
chapter of the ADSP-2126x Hardware Reference Manual. Note  
Table 27. Parallel Data Acquisition Port (PDAP)  
Parameter  
Min  
Max  
Unit  
Timing Requirements  
tSPCLKEN  
tHPCLKEN  
tPDSD  
PDAP_CLKEN Setup Before PDAP_CLK Sample Edge1  
PDAP_CLKEN Hold After PDAP_CLK Sample Edge1  
PDAP_DAT Setup Before SCLK PDAP_CLK Sample Edge1  
PDAP_DAT Hold After SCLK PDAP_CLK Sample Edge1  
Clock Width  
2.5  
2.5  
2.5  
2.5  
7
ns  
ns  
ns  
ns  
ns  
ns  
tPDHD  
tPDCLKW  
tPDCLK  
Clock Period  
20  
Switching Characteristics  
tPDHLDD Delay of PDAP Strobe After Last PDAP_CLK Capture Edge for a Word  
tPDSTRB PDAP Strobe Pulse Width  
2 x tCCLK  
ns  
ns  
1 x tCCLK – 1  
1
Source pins of DATA are ADDR[7:0], DATA[7:0], or DAI pins. Source pins for SCLK and FS are: 1) DAI pins, 2) CLKIN through PCG, or 3) DAI pins through PCG.  
SAMPLE EDGE  
tPDCLKW  
DAI_P[20:1]  
(PDAP_CLK)  
tSPCLKEN  
tHPCLKEN  
DAI_P[20:1]  
(PDAP_CLKEN)  
tPDSD  
tPDHD  
DATA  
DAI_P[20:1]  
(PDAP_STROBE)  
tPDSTRB  
tPDHLDD  
Figure 24. PDAP Timing  
Rev. PrA  
|
Page 32 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
SPI Interface—Master  
Table 28. SPI Interface Protocol — Master Switching and Timing Specifications  
Parameter  
Min  
Max  
Unit  
Timing Requirements  
tSSPIDM  
tHSPIDM  
Data input valid to SPICLK edge (data input set-up time)  
SPICLK Last Sampling Edge to Data Input Not Valid  
5
2
ns  
ns  
Switching Characteristics  
tSPICLKM  
tSPICHM  
tSPICLM  
tDDSPIDM  
tHDSPIDM  
tSDSCIM  
tHDSM  
Serial Clock Cycle  
8 x tCCLK  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
SeriaL Clock High Period  
4 x tCCLK – 2  
4 x tCCLK – 2  
SeriaL Clock Low Period  
SPICLK Edge to Data Out Valid (Data Out Delay Time)  
SPICLK Edge to Data Out Not Valid (Data Out Hold Time)  
FLAG3-0 OUT (SPI Device Select) Low to First SPICLK Edge  
Last SPICLK Edge to FLAG3-0 OUT High  
Sequential Transfer Delay  
3
10  
4 x tCCLK – 2  
4 x tCCLK – 1  
4 x tCCLK – 1  
tSPITDM  
FLG3-0  
(OUTPUT)  
tSD SCIM  
tSPICH M  
tSPIC LM  
tSPICHM  
tDDSPIDM  
tSPIC LK M  
tHDSM  
tSPITDM  
SPICLK  
(CP = 0)  
(OUTPUT)  
tSPIC LM  
SPICLK  
(CP = 1)  
(OUTPUT)  
tHDSPIDM  
MOSI  
(OUTPUT)  
MSB  
LSB  
tSSPID M  
tSSPIDM  
CPHASE = 1  
tHSPIDM  
tHSPIDM  
MISO  
MSB  
LSB  
(INPUT)  
VALID  
VALID  
tHDSPIDM  
tDDSPIDM  
MOSI  
(OUTPUT)  
MSB  
LSB  
tSSPIDM  
tH SPID M  
CPHASE = 0  
MSB  
VALID  
LSB  
VALID  
MISO  
(INPUT)  
Figure 25. SPI Master Timing  
Rev. PrA  
|
Page 33 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
SPI Interface—Slave  
Table 29. SPI Interface Protocol —Slave Switching and Timing Specifications  
Parameter  
Min  
Max  
Unit  
Timing Requirements  
tSPICLKS  
tSPICHS  
tSPICLS  
tSDSCO  
Serial Clock Cycle  
4 x tCCLK  
ns  
ns  
ns  
Serial Clock High Period  
Serial Clock Low Period  
2 x tCCLK – 2  
2 x tCCLK – 2  
SPIDS Assertion to First SPICLK Edge  
CPHASE = 0  
CPHASE = 1  
2 x tCCLK + 1  
2 x tCCLK + 1  
ns  
ns  
tHDS  
Last SPICLK Edge to SPIDS Not Asserted CPHASE = 0  
Data Input Valid to SPICLK Edge (Data Input Set-up Time)  
SPICLK Last Sampling Edge to Data Input Not Valid  
SPIDS Deassertion Pulse Width (CPHASE=0)  
2 x tCCLK  
ns  
ns  
ns  
ns  
tSSPIDS  
tHSPIDS  
tSDPPW  
2
2
2 x tCCLK  
Switching Characteristics  
tDSOE  
SPIDS Assertion to Data Out Active  
0
0
5
ns  
ns  
ns  
ns  
ns  
tDSDHI  
tDDSPIDS  
tHDSPIDS  
tDSOV  
SPIDS Deassertion to Data High Impedance  
5
SPICLK Edge to Data Out Valid (Data Out Delay Time)  
SPICLK Edge to Data Out Not Valid (Data Out Hold Time)  
SPIDS Assertion to Data Out Valid (CPHASE=0)  
7.5  
2 x tCCLK – 2  
5 x tCCLK + 2  
SPIDS  
(INPUT)  
tSPIC HS  
tSPICLS  
tSPICLKS  
tHDS  
tSDPPW  
SPICLK  
(CP = 0)  
(INPUT)  
tSPICLS  
tSDSCO  
tSPICHS  
SPICLK  
(CP = 1)  
(INPUT)  
tDSDHI  
tHDSPIDS  
tDDSPIDS  
tDSOE  
tDDSPIDS  
MISO  
(OUTPUT)  
MSB  
LSB  
tHSPIDS  
CPHASE = 1  
tSSPIDS  
tSSPIDS  
MOSI  
(INPUT)  
LSB VALID  
MSB VALID  
tDSOV  
tDSO E  
tHDSPIDS  
tDDSPIDS  
tDSDHI  
MISO  
(OUTPUT)  
LSB  
MSB  
tHSPIDS  
CPHASE = 0  
tSSPIDS  
MOSI  
MSB VALID  
LSB VALID  
(INPUT)  
Figure 26. SPI Slave Timing  
Rev. PrA  
|
Page 34 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
JTAG Test Access Port and Emulation  
Table 30. JTAG Test Access Port and Emulation  
Parameter  
Min  
Max  
Unit  
Timing Requirements  
tTCK  
TCK Period  
20  
5
ns  
ns  
ns  
ns  
ns  
ns  
tSTAP  
tHTAP  
tSSYS  
tHSYS  
tTRSTW  
TDI, TMS Setup Before TCK High  
TDI, TMS Hold After TCK High  
System Inputs Setup Before TCK Low1  
System Inputs Hold After TCK Low1  
TRST Pulse Width  
6
7
8
4tCK  
Switching Characteristics  
tDTDO TDO Delay from TCK Low  
tDSYS  
System Outputs Delay After TCK Low2  
7
ns  
ns  
10  
1 System Inputs = AD15-0, SPIDS, CLKCFG1-0, RESET, BOOTCFG1-0, MISO, MOSI, SPICLK, DAI_Px, FLAG3-0  
2 System Outputs = MISO, MOSI, SPICLK, DAI_Px, AD15-0, RD, WR, FLAG3-0, CLKOUT, EMU, ALE.  
tTCK  
TCK  
tSTAP  
tHTAP  
TMS  
TDI  
tDTDO  
TDO  
tSSYS  
tHSYS  
SYSTEM  
INPUTS  
tDSYS  
SYSTEM  
OUTPUTS  
Figure 27. IEEE 11499.1 JTAG Test Access Port  
Rev. PrA  
|
Page 35 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
OUTPUT DRIVE CURRENTS  
CAPACITIVE LOADING  
Figure 28 shows typical I-V characteristics for the output driv-  
ers of the ADSP-21267. The curves represent the current drive  
capability of the output drivers as a function of output voltage.  
Output delays and holds are based on standard capacitive loads:  
12 pF on all pins (see Figure 29). Figure 32 shows graphically  
how output delays and holds vary with load capacitance (Note  
that this graph or derating does not apply to output disable  
delays. The graphs of Figure 31, Figure 32 and Figure 33 may  
not be linear outside the ranges shown for Typical Output Delay  
vs. Load Capacitance and Typical Output Rise Time (20%-80%,  
V=Min) vs. Load Capacitance.  
40  
V
OH TBD  
30  
20  
10  
0
12.0  
10.0  
8.0  
-10  
-20  
TBD  
V
-30  
-40  
OL TBD  
0.5  
6.0  
0
1
1.5  
2
2.5  
3
3.5  
4.0  
2.0  
0
SWEEP (V  
) VOLTAGE - V  
DDEXT  
Figure 28. ADSP-21267 Typical Drive  
TEST CONDITIONS  
0
20  
40  
60  
80  
100  
120  
LOAD CAPACITANCE - PF  
The ac signal specifications (timing parameters) appear Table 9  
on page 18 through Table 30 on page 35. These include output  
disable time, output enable time, and capacitive loading.  
Figure 31. Typical Output Rise Time  
(20%-80%, VDDEXT = Max)  
Timing is measured on signals when they cross the 1.5 V level as  
described in Figure 30 on page 36. All delays (in nanoseconds)  
are measured between the point that the first signal reaches  
1.5 V and the point that the second signal reaches 1.5 V.  
12  
10  
8
50  
TO  
1.5V  
OUTPUT  
PIN  
6
4
2
0
TBD  
30pF  
Figure 29. Equivalent Device Loading for AC Measurements  
(Includes All Fixtures)  
0
20  
40  
60  
80  
100  
120  
LOAD CAPACITANCE - pF  
INPUT  
OR  
OUTPUT  
1.5V  
1.5V  
Figure 32. Typical Output Rise/Fall Time  
(20%-80%, VDDEXT = Min)  
Figure 30. Voltage Reference Levels for AC Measurements  
Rev. PrA  
|
Page 36 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Where:  
TA = Ambient Temperature 0C  
7
6
5
4
3
2
Values of θJC are provided for package comparison and PCB  
design considerations when an external heatsink is required.  
Values of θJB are provided for package comparison and PCB  
design considerations.  
TBD  
Table 31. Thermal Characteristics for 136 Ball BGA  
1
0
Parameter  
θJA  
Condition  
Typical  
28.2  
24.4  
23.3  
20.1  
7.0  
Unit  
°C/W  
°C/W  
°C/W  
°C/W  
°C/W  
°C/W  
°C/W  
°C/W  
-1  
-2  
-3  
-4  
Airflow = 0 m/s  
Airflow = 1 m/s  
Airflow = 2 m/s  
θJMA  
θJMA  
0
20  
40  
60  
80  
100  
120  
θJB  
θJC  
LOAD CAPACITANCE - pF  
ΨJT  
Airflow = 0 m/s  
Airflow = 1 m/s  
Airflow = 2 m/s  
0.1  
ΨJMT  
ΨJMT  
0.3  
Figure 33. Typical Output Delay or Hold vs. Load Capacitance  
(at Ambient Temperature)  
0.4  
ENVIRONMENTAL CONDITIONS  
The ADSP-21267 processor is rated for performance over the  
commercial temperature range, TAMB = 0°C to 70°C.  
Table 32. Thermal Characteristics for 144 Lead LQFP  
Parameter  
θJA  
Typical  
32.5  
28.9  
27.8  
7.8  
Unit  
°C/W  
°C/W  
°C/W  
°C/W  
°C/W  
°C/W  
°C/W  
THERMAL CHARACTERISTICS  
Airflow = 0 m/s  
Airflow = 1 m/s  
Airflow = 2 m/s  
The ADSP-21267 is offered in 144-lead LQFP and 136-ball BGA  
packages  
θJMA  
θJMA  
Table 31 and Table 32 airflow measurements comply with  
JEDEC standards JESD51-2 and JESD51-6 and the junction-to-  
board measurement complies with JESD51-8. The junction-to-  
case measurement complies with MIL- STD-883. All measure-  
ments use a 2S2P JEDEC test board.  
θJC  
ΨJT  
Airflow = 0 m/s  
Airflow = 1 m/s  
Airflow = 2 m/s  
0.5  
ΨJMT  
ΨJMT  
0.8  
1.0  
To determine the Junction Temperature of the device while on  
the application PCB, use:  
T
= T  
+ (Ψ × PD)  
CASE JT  
J
Where:  
TJ= Junction temperature 0C  
TCASE= Case temperature (0C) measured at the top center of the  
package  
ΨJT= Junction-to-Top (of package) characterization parameter  
= Typical value from the tables below  
PD= Power dissipation see EE Note #216  
Values of θJA are provided for package comparison and PCB  
design considerations. θJA can be used for a 1st order approxima-  
tion of TJ by the equation:  
T
= T + (θ × PD)  
JA  
J
A
Rev. PrA  
|
Page 37 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
136-BALL BGA PIN CONFIGURATIONS  
The following table shows the ADSP-21267’s pin names and  
their default function after reset (in parenthesis).  
Table 33. 136-ball BGA Pin Assignments  
Pin Name  
BGA  
Pin#  
Pin Name  
BGA  
Pin#  
Pin Name  
BGA  
Pin#  
Pin Name  
BGA  
Pin#  
CLKCFG0  
XTAL  
TMS  
A01  
A02  
A03  
A04  
A05  
A06  
A07  
A08  
A09  
A10  
A11  
A12  
A13  
A14  
E01  
E02  
E04  
E05  
E06  
E09  
E10  
E11  
E13  
E14  
CLKCFG1  
GND  
B01  
B02  
B03  
B04  
B05  
B06  
B07  
B08  
B09  
B10  
B11  
B12  
B13  
B14  
F01  
F02  
F04  
F05  
F06  
F09  
F10  
F11  
F13  
F14  
BOOTCFG1  
BOOTCFG0  
GND  
C01  
C02  
C03  
C12  
C13  
C14  
VDDINT  
GND  
GND  
GND  
GND  
GND  
GND  
GND  
GND  
VDDINT  
D01  
D02  
D04  
D05  
D06  
D09  
D10  
D11  
D13  
D14  
VDDEXT  
CLKIN  
TRST  
TCK  
GND  
TDI  
GND  
CLKOUT  
TDO  
AVSS  
VDDINT  
AVDD  
EMU  
VDDEXT  
SPICLK  
RESET  
VDDINT  
GND  
MOSI  
MISO  
SPIDS  
VDDINT  
GND  
GND  
VDDINT  
GND  
GND  
GND  
GND  
GND  
GND  
GND  
GND  
FLAG3  
GND  
GND  
FLAG1  
FLAG0  
GND  
AD7  
G01  
G02  
G13  
G14  
AD6  
H01  
H02  
H13  
H14  
VDDINT  
VDDEXT  
VDDEXT  
DAI_P18 (SD5B)  
DAI_P17 (SD5A)  
GND  
DAI_P19 (SCLK45)  
GND  
GND  
GND  
GND  
FLAG2  
DAI_P20 (SFS45)  
Rev. PrA  
|
Page 38 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Table 33. 136-ball BGA Pin Assignments (Continued)  
Pin Name  
BGA  
Pin#  
Pin Name  
BGA  
Pin#  
Pin Name  
BGA  
Pin#  
Pin Name  
BGA  
Pin#  
AD5  
J01  
AD3  
K01  
K02  
K04  
K05  
K06  
K09  
K10  
K11  
K13  
K14  
P01  
P02  
P03  
P04  
P05  
P06  
P07  
P08  
P09  
P10  
P11  
P12  
P13  
P14  
AD2  
L01  
L02  
L04  
L05  
L06  
L09  
L10  
L11  
L13  
L14  
AD0  
WR  
M01  
M02  
M03  
M12  
AD4  
J02  
VDDINT  
AD1  
GND  
J04  
GND  
GND  
GND  
GND  
GND  
J05  
GND  
GND  
GND  
J06  
GND  
GND  
DAI_P12 (SD3B)  
M13  
M14  
GND  
J09  
GND  
GND  
DAI_P13 (SCLK23)  
GND  
J10  
GND  
GND  
GND  
J11  
GND  
GND  
VDDINT  
J13  
GND  
GND  
DAI_P16 (SD4B)  
AD15  
J14  
DAI_P15 (SD4A)  
AD14  
DAI_P14 (SFS23)  
N01  
N02  
N03  
N04  
N05  
N06  
N07  
N08  
N09  
N10  
N11  
N12  
N13  
N14  
ALE  
AD13  
RD  
AD12  
VDDINT  
AD11  
VDDEXT  
AD10  
AD8  
AD9  
VDDINT  
DAI_P1 (SD0A)  
DAI_P3 (SCLK0)  
DAI_P5 (SD1A)  
DAI_P6 (SD1B)  
DAI_P7 (SCLK1)  
DAI_P8 (SFS1)  
DAI_P9 (SD2A)  
DAI_P11 (SD3A)  
DAI_P2 (SD0B)  
VDDEXT  
DAI_P4 (SFS0)  
VDDINT  
VDDINT  
GND  
DAI_P10 (SD2B)  
Rev. PrA  
|
Page 39 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
14 13 12 11 10  
9
8
7
6
5
4
3
2
1
A
B
C
D
E
F
G
H
J
K
L
M
N
P
KEY  
V
A
VDD  
GND*  
DDINT  
V
A
I/O SIGNALS  
DDEXT  
VSS  
*USE THE CENTER BLOCK OF GROUND PINS TO PROVIDE  
THERMAL PATHWAYS TO YOUR PRINTED  
CIRCUIT BOARD’S GROUND PLANE.  
Figure 34. 136-ball BGA Pin Assignments (Bottom View, Summary)  
Rev. PrA  
|
Page 40 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
144-LEAD LQFP PIN CONFIGURATIONS  
The following table shows the ADSP-21267’s pin names and  
their default function after reset (in parenthesis).  
Table 34. 144-Lead LQFP Pin Assignments  
Pin Name  
LQFP  
Pin #  
Pin Name  
LQFP  
Pin #  
Pin Name  
LQFP  
Pin #  
Pin Name  
LQFP  
Pin #  
VDDINT  
CLKCFG0  
CLKCFG1  
BOOTCFG0  
BOOTCFG1  
GND  
1
VDDINT  
GND  
RD  
37  
38  
39  
40  
41  
42  
43  
44  
45  
46  
47  
48  
49  
50  
51  
52  
VDDEXT  
73  
74  
75  
76  
77  
78  
79  
GND  
VDDINT  
GND  
VDDINT  
GND  
VDDINT  
GND  
VDDEXT  
GND  
VDDINT  
GND  
VDDINT  
RESET  
SPIDS  
GND  
VDDINT  
SPICLK  
MISO  
MOSI  
GND  
VDDINT  
VDDEXT  
AVDD  
109  
110  
111  
112  
113  
114  
115  
116  
117  
118  
119  
120  
121  
122  
123  
124  
125  
126  
127  
128  
129  
130  
131  
132  
133  
134  
135  
136  
137  
138  
139  
140  
141  
142  
143  
144  
2
GND  
3
VDDINT  
4
ALE  
GND  
5
AD15  
AD14  
AD13  
GND  
VDDEXT  
AD12  
VDDINT  
GND  
AD11  
AD10  
AD9  
DAI_P10 (SD2B)  
DAI_P11 (SD3A)  
DAI_P12 (SD3B)  
6
VDDEXT  
GND  
7
8
DAI_P13 (SCLK23) 80  
VDDINT  
GND  
9
DAI_P14 (SFS23)  
DAI_P15 (SD4A)  
VDDINT  
81  
82  
83  
84  
85  
86  
87  
88  
10  
11  
12  
13  
14  
15  
16  
17  
18  
19  
20  
21  
22  
23  
24  
25  
26  
27  
28  
29  
30  
31  
32  
33  
34  
35  
36  
VDDINT  
GND  
GND  
VDDINT  
GND  
GND  
DAI_P16 (SD4B)  
DAI_P17 (SD5A)  
DAI_P18 (SD5B)  
FLAG0  
FLAG1  
AD7  
AD8  
DAI_P1 (SD0A) 53  
DAI_P19 (SCLK45) 89  
GND  
VDDINT  
GND  
54  
55  
VDDINT  
GND  
90  
VDDINT  
GND  
91  
DAI_P2 (SD0B) 56  
DAI_P3 (SCLK0) 57  
GND  
92  
VDDEXT  
GND  
VDDEXT  
DAI_P20 (SFS45)  
GND  
93  
GND  
VDDEXT  
VDDINT  
GND  
58  
59  
60  
61  
94  
VDDINT  
AD6  
95  
VDDINT  
FLAG2  
FLAG3  
VDDINT  
GND  
96  
AVSS  
AD5  
97  
GND  
CLKOUT  
EMU  
TDO  
AD4  
DAI_P4 (SFS0) 62  
DAI_P5 (SD1A) 63  
DAI_P6 (SD1B) 64  
DAI_P7 (SCLK1) 65  
98  
VDDINT  
GND  
99  
100  
101  
102  
103  
104  
105  
106  
107  
108  
AD3  
VDDINT  
GND  
TDI  
AD2  
VDDINT  
GND  
VDDINT  
GND  
66  
67  
68  
69  
TRST  
TCK  
VDDEXT  
GND  
VDDINT  
GND  
TMS  
AD1  
VDDINT  
GND  
GND  
CLKIN  
XTAL  
VDDEXT  
AD0  
DAI_P8 (SFS1) 70  
DAI_P9 (SD2A) 71  
WR  
VDDINT  
VDDINT  
VDDINT  
VDDINT  
72  
Rev. PrA  
|
Page 41 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
PACKAGE DIMENSIONS  
The ADSP-21267 is available in a 136-ball BGA package and a  
144-lead LQFP package. All dimensions are in millimeters  
(mm).  
Figure 35. 136-ball BGA  
10.40  
BSC  
12.00  
SQ  
BSC  
A1 BALL  
PAD CORNER  
0.80  
TYP  
A
B
C
D
E
F
G
H
J
A1 BALL  
PAD CORNER  
12.00  
SQ  
BSC  
10.40  
BSC  
Top View  
K
L
M
N
P
14 13 12 11 10 9  
8 7 6 5 4 3 2 1  
0.80  
TYP  
DETAIL A  
1.70  
MAX  
DETAIL A  
1.31  
1.21  
1.10  
ALL DIMENSIONS IN MILIMETERS (MM).  
0.25  
MIN  
1. THE ACTUAL POSITION OF THE BALL GRID IS  
WITHIN 0.150 MM OF ITS IDEAL POSITION RELATIVE  
TO THE PACKAGE EDGES. THE ACTUAL POSITION  
OF EACH BALL IS WITHIN 0.08 MM OF ITS IDEAL  
POSITION RELATIVE TO THE BALL GRID.  
SEATING  
PLANE  
0.50  
0.46  
0.40  
0.12  
MAX  
2. COMPLIANT TO JEDEC REGISTERED OUTLINE  
MO-205-AE WITH THE EXCEPTION OF DIMENSION “b”  
BALL  
DIAMETER  
Rev. PrA  
|
Page 42 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
Figure 36. 144-lead LQFP (ST-144)  
22.00 BSC SQ  
20.00 BSC SQ  
109  
144  
1
108  
PIN 1 INDICATOR  
0.50  
BSC  
TYP  
(LEAD  
PITCH)  
0.27  
0.22  
0.17  
TYP  
SEATING  
PLANE  
0.08 MAX (LEAD  
COPLANARITY)  
NOTES:  
1. DIMENSIONS ARE IN MILLIMETERS AND COMPLY  
WITH JEDEC STANDARD MS-026-BFB.  
0.15  
0.05  
2. ACTUAL POSITION OF EACH LEAD IS WITHIN 0.08  
OF ITS IDEAL POSITION, WHEN MEASURED IN THE  
LATERAL DIRECTION.  
1.45  
1.40  
1.35  
0.75  
3. CENTER DIMENSIONS ARE NOMINAL.  
0.60 TYP  
0.45  
73  
3 6  
72  
37  
1.60 MAX  
DETAIL A  
DETAIL A  
TOP VIEW (PINS DOWN)  
ORDERING GUIDE  
Analog Devices offers a wide variety of audio algorithms and  
combinations to run on the ADSP-21267 DSP. These products  
are sold as part of a chip set, bundled with necessary application  
software under special part numbers. For a complete list, visit  
our web site at www.analog.com\SHARC.  
These product also may contain 3rd party IPs that may require  
users to have authorization from the respective IP holders to  
receive them. Royalty for use of the 3rd party IPs may also be  
payable by users.  
Part Number1,2,3  
Ambient Temper- InstructionRate On-Chip  
ROM  
Operating Voltage Package  
ature Range  
0°C to +70°C  
0°C to +70°C  
SRAM  
1 Mbit  
1 Mbit  
ADSP-21267SKSTZ-X  
ADSP-21267SKBCZ-X  
150 MHz  
150 MHz  
3 Mbit  
3 Mbit  
1.2 INT/3.3 EXT V  
1.2 INT/3.3 EXT V  
144-Lead LQFP  
136-Lead BGA  
1 K indicates commercial grade temperature (0°C to +70°C).  
2 B indicates Ball Grid Array package. ST indicates Low Profile Quad Flat package.  
3 Z indicates Lead Free package. For more information about lead free package offerings, please visit www.analog.com.  
Rev. PrA  
|
Page 43 of 44  
|
January 2004  
PRELIMINARY TECHNICAL DATA  
ADSP-21267  
© 2004 Analog Devices, Inc. All rights reserved. Trademarks and  
registered trademarks are the property of their respective owners.  
www.analog.com  
a
PR04623-0-1/04(PrA)  
Rev. PrA  
|
Page 44 of 44  
|
January 2004  

相关型号:

ADSP-21267SKSTZ-X

Preliminary Technical Data
ADI

ADSP-21362

SHARC Processor
ADI

ADSP-21362BBC-1AA

SHARC Processor
ADI

ADSP-21362BBCZ-1AA

SHARC Processor
ADI

ADSP-21362BSQZ-1AA

IC 16-BIT, 55.55 MHz, OTHER DSP, PQFP144, LEAD FREE, MS-026BFB-HD, LQFP-144, Digital Signal Processor
ADI

ADSP-21362BSWZ-1AA

SHARC Processors
ADI

ADSP-21362KBC-1AA

SHARC Processor
ADI

ADSP-21362KBC-1AX

IC 16-BIT, 55.55 MHz, OTHER DSP, PBGA136, MO-205AE, BGA-136, Digital Signal Processor
ADI

ADSP-21362KBCZ-1AA

SHARC Processor
ADI

ADSP-21362KBCZ-1AX

IC 16-BIT, 55.55 MHz, OTHER DSP, PBGA136, LEAD FREE, MO-205AE, BGA-136, Digital Signal Processor
ADI

ADSP-21362KSQZ-1AA

IC 16-BIT, 55.55 MHz, OTHER DSP, PQFP144, LEAD FREE, MS-026BFB-HD, LQFP-144, Digital Signal Processor
ADI

ADSP-21362KSQZ-1AX

IC 16-BIT, 55.55 MHz, OTHER DSP, PQFP144, LEAD FREE, MS-026BFB-HD, HSLQFP-144, Digital Signal Processor
ADI